RADIOSS Element

RADIOSS THEORY MANUAL
Version 2017 – January 2017

Large Displacement Finite Element Analysis
Chapter 5
Altair Engineering, Inc., World Headquarters: 1820 E. Big Beaver Rd., Troy MI 48083-2031 USA
Phone: +1.248.614.2400 • Fax: +1.248.614.2411 • www.altair.com • info@altair.com
RADIOSS THEORY Version 2017 CONTENTS
CONTENTS
5.1 SOLID HEXAHEDRON ELEMENTS 5
5.1.1 SHAPE FUNCTIONS FOR LINEAR BRICKS 5
5.1.2 STRAIN RATE 7
5.1.3 ASSUMED STRAIN RATE 7
5.1.4 INTERNAL FORCE CALCULATION 9
5.1.5 HOURGLASS MODES 10
5.1.6 STABILITY 14
5.1.7 SHOCK WAVES 15
5.1.8 ELEMENT DEGENERATION 15
5.1.9 INTERNAL STRESS CALCULATION 17
5.1.10 DEVIATORIC STRESS CALCULATION 20
5.2 SOLID TETRAHEDRON ELEMENTS 23
5.2.1 4-NODE SOLID TETRAHEDRON 23
5.2.2 10-NODE SOLID TETRAHEDRON 23
5.3 SHELL ELEMENTS 27
5.3.1 INTRODUCTION 27
5.3.2 BILINEAR MINDLIN PLATE ELEMENT 28
5.3.3 TIME STEP FOR STABILITY 29
5.3.4 LOCAL REFERENCE FRAME 29
5.3.5 BILINEAR SHAPE FUNCTIONS 30
5.3.6 MECHANICAL PROPERTIES 31
5.3.7 INTERNAL FORCES 35
5.3.8 HOURGLASS MODES 35
5.3.9 HOURGLASS RESISTANCE 37
5.3.10 STRESS AND STRAIN CALCULATION 41
5.3.11 CALCULATION OF FORCES AND MOMENTS 48
5.3.12 QPH, QPPS, QEPH AND QBAT SHELL FORMULATIONS 50
5.3.12.1 FORMULATIONS FOR A GENERAL DEGENERATED 4-NODE SHELL 51
5.3.12.2 FULLY-INTEGRATED SHELL ELEMENT QBAT 54
5.3.12.3 THE NEW ONE-POINT QUADRATURE SHELL ELEMENT 54
5.3.12.4 ADVANCED ELASTO-PLASTIC HOURGLASS CONTROL 60
5.3.13 THREE-NODE SHELL ELEMENTS 61
5.3.14 COMPOSITE SHELL ELEMENTS 64
5.3.14.1 TRANSFORMATION MATRIX FROM GLOBAL TO ORTHOTROPIC SKEW 64
5.3.14.2 COMPOSITE MODELING IN RADIOSS 65
5.3.14.3 ELEMENT ORIENTATION 66
5.3.14.4 ORTHOTROPIC SHELLS 67
5.3.14.5 COMPOSITE SHELL 67
5.3.14.6 COMPOSITE SHELL WITH VARIABLE LAYERS 68
5.3.14.7 LIMITATIONS 68
5.3.15 THREE-NODE TRIANGLE WITHOUT ROTATIONAL D.O.F. (SH3N6) 68
5.3.15.1 STRAIN COMPUTATION 69
5.3.15.2 BOUNDARY CONDITIONS APPLICATION 71
5.4 SOLID-SHELL ELEMENTS 73
5.5 TRUSS ELEMENTS (TYPE 2) 76
5.5.1 PROPERTY INPUT 76
5.5.2 STABILITY 76
5.5.3 RIGID BODY MOTION 76
5.5.4 STRAIN 77
5.5.5 MATERIAL TYPE 77
5.5.6 FORCE CALCULATION 77
5.6 BEAM ELEMENTS (TYPE 3) 78
5.6.1 LOCAL COORDINATE SYSTEM 78
5.6.2 BEAM ELEMENT GEOMETRY 78
01-Jan-2017 1
5.6.3 MINIMUM TIME STEP 79

5.6.4 BEAM ELEMENT BEHAVIOR 79
5.6.5 MATERIAL PROPERTIES 82
5.6.6 INERTIA COMPUTATION 84
5.7 ONE DEGREE OF FREEDOM SPRING ELEMENTS (TYPE 4) 85
5.7.1 TIME STEP 86
5.7.2 LINEAR SPRING 86
5.7.3 NONLINEAR ELASTIC SPRING 87
5.7.4 NONLINEAR ELASTO-PLASTIC SPRING - ISOTROPIC HARDENING 88
5.7.5 NONLINEAR ELASTO-PLASTIC SPRING - DECOUPLED HARDENING 89
5.7.6 NONLINEAR ELASTIC-PLASTIC SPRING - KINEMATIC HARDENING 89
5.7.7 NONLINEAR ELASTO-PLASTIC SPRING - NON LINEAR UNLOADING 90
5.7.8 NONLINEAR DASHPOT 90
5.7.9 NONLINEAR VISCOELASTIC SPRING 91
5.8 GENERAL SPRING ELEMENTS (TYPE 8) 93
5.8.1 TIME STEP 93
5.8.6 NONLINEAR ELASTO-PLASTIC SPRING - KINEMATIC HARDENING 93
5.8.9 NONLINEAR VISCOELASTIC SPRING 93
5.8.10 SKEW FRAME PROPERTIES 93
5.8.11 DEFORMATION SIGN CONVENTION 95
5.8.12 TRANSLATIONAL FORCES 95
5.8.13 MOMENTS 96
5.8.14 MULTIDIRECTIONAL FAILURE CRITERIA 97
5.9 PULLEY TYPE SPRING ELEMENTS (TYPE 12) 98
5.9.1 TIME STEP 98
5.9.7 NONLINEAR VISCO-ELASTIC SPRING 99
5.9.8 FRICTION EFFECTS 99
5.10 BEAM TYPE SPRING ELEMENTS (TYPE 13) 101
5.10.1 TIME STEP 101
5.10.6 NONLINEAR ELASTO-PLASTIC SPRING - KINEMATIC HARDENING 102
5.10.9 NONLINEAR VISCO-ELASTIC SPRING 102
5.10.10 SKEW FRAME PROPERTIES 102
5.10.11 SIGN CONVENTIONS 103
5.10.12 TENSION 103
5.10.13 SHEAR - XY 104
5.10.14 SHEAR - XZ 105
5.10.15 TORSION 106
5.10.16 BENDING ABOUT THE Y AXIS 106
5.10.17 BENDING ABOUT THE Z AXIS 106
5.10.18 MULTIDIRECTIONAL FAILURE CRITERIA 107
01-Jan-2017 2
5.11 MULTISTRAND ELEMENTS (TYPE 28) 108

5.11.1 INTRODUCING MULTISTRAND ELEMENTS 108
5.11.2 INTERNAL FORCES COMPUTATION 108
5.12 SPRING TYPE PRETENSIONER (TYPE 32) 111
5.12.1 INTRODUCING PRETENSIONERS 111
5.12.2 RADIOSS MODEL FOR PRETENSIONERS 111
01-Jan-2017 3
Chapter
5
ELEMENT LIBRARY
01-Jan-2017 4
RADIOSS THEORY Version 2017 ELEMENT LIBRARY
5.0 ELEMENT LIBRARY

RADIOSS element library contains elements for one, two or three dimensional problems. Some new elements
have been developed and implemented in recent versions. Most of them use the assumed strain method to avoid
some locking problems. For the elements using reduced integration schema, the physical stabilization method is
used to control efficiently the hourglass deformations. Another point in these new elements is the use of co-
rotational coordinate system. For the new solid elements, as the assumed strains are often defined in the specific
directions, the use of global system combined with Jaumman's stress derivation contributes to commutative error
especially when solid undergoes large shear strains.
The RADIOSS finite element library can be classified into the following categories of elements:
• Solid elements : 8- and 20-node bricks, 4- and 10-node tetrahedrons,

• Solid-shells : 8- , 16- and 20-node hexahedrons, 6-node pentahedral element,
• 2 dimensional elements : 4-node quadrilaterals for plane strain and axisymmetrical analysis,
• Shell elements : 4-node quadrilaterals and 3-node triangles,
• One dimensional elements: rivet, springs, bar and beams.
The implementation of these elements will now be detailed. Expression of nodal forces will be developed as, for
explicit codes they represent the discretization of the momentum equations. Stiffness matrices, which are central
to implicit finite element approaches, are not developed here.
5.1 Solid Hexahedron Elements

RADIOSS brick elements have the following properties:
• BRICK8: 8-node linear element with reduced or full integration,
• HA8: 8-node linear element with various number of integration points going from 2x2x2 to 9x9x9,
• HEPH: 8-node linear element with reduced integration point and physical stabilization of hourglass
modes,
• BRICK20: 20-node quadratic element with reduced or full integration schemes.
For all elements, a lumped mass approach is used and the elements are isoparametric, i.e. the same shape
functions are used to define element geometry and element displacements
The fundamental theory of each element is described in this chapter.
5.1.1 Shape functions for linear bricks

Shape functions define the geometry of an element in its computational (intrinsic) domain. As was seen in
Chapter 3, physical coordinates are transformed into simpler computational intrinsic coordinates so that
integration of values is numerically more efficient.
01-Jan-2017 5
Figure 5.1.1 8 Node Brick Element
Where: r ≡ ξ , s ≡η , t ≡ ζ
The shape functions of an 8 node brick element, shown in Figure 5.1.1, are given by:
Φ1 =
1
(1 − ξ )(1 − η )(1 − ζ ) EQ. 5.1.1.1
8
Φ2 =
1
(1 − ξ )(1 − η )(1 + ζ ) EQ. 5.1.1.2
8
Φ3 =
1
(1 + ξ )(1 −η )(1 + ζ ) EQ. 5.1.1.3
8
Φ4 =
1
(1 + ξ )(1 − η )(1 − ζ ) EQ. 5.1.1.4
8
Φ5 =
1
(1 − ξ )(1 + η )(1 − ζ ) EQ. 5.1.1.5
8
Φ6 =
1
(1 − ξ )(1 + η )(1 + ζ ) EQ. 5.1.1.6
8
Φ7 =
1
(1 + ξ )(1 + η )(1 + ζ ) EQ. 5.1.1.7
8
Φ8 =
1
(1 + ξ )(1 + η )(1 − ζ ) EQ. 5.1.1.8
8
The element velocity field is related by:
8
vi = ∑ Φ I .viI EQ. 5.1.1.9
I =1
where the viI are the nodal velocities.
01-Jan-2017 6
5.1.2 Strain rate

The relationship between the physical coordinate and computational intrinsic coordinates system for a brick
element is given by the matrix equation:
 ∂Φ I   ∂x ∂y ∂z   ∂Φ I   ∂Φ I 
 ∂ξ   ∂ξ ∂ξ ∂ξ   ∂x   ∂x 
      ∂Φ 
 ∂Φ I  =  ∂x ∂y ∂z   ∂Φ I 
. = Fξ . I  EQ. 5.1.2.1
 ∂η   ∂η ∂η ∂η   ∂y   ∂y 
 ∂Φ   ∂x ∂y ∂z   ∂Φ I   ∂Φ I 
 I     
 ∂ζ   ∂ζ ∂ζ ∂ζ   ∂z   ∂z 
Hence:
 ∂Φ I  −1  ∂Φ I 
  = Fξ .  EQ. 5.1.2.2
 ∂xi   ∂ξ 
where Fξ is the Jacobian matrix.

The element strain rate is defined as:
1  ∂v ∂v 
εɺij =  i + j  EQ. 5.1.2.3
2  ∂x j ∂xi 
Relating the element velocity field to its shape function gives:
∂vi 8
∂Φ I
=∑ ⋅ viI EQ. 5.1.2.4
∂x j I =1 ∂x j
Hence, the strain rate can be described directly in terms of the shape function:

1 ∂v ∂vj  8 ∂ΦI
εɺij =  i + =∑ ⋅ viI EQ. 5.1.2.5
2  ∂xj ∂xi  I =1 ∂xj
As was seen in section 2.4.1, volumetric strain rate is calculated separately by volume variation.
For one integration point:
∂Φ1 ∂Φ 7 ∂Φ 2 ∂Φ 8 ∂Φ 3 ∂Φ 5 ∂Φ 4 ∂Φ 6
=− ; =− ; =− ; =− EQ. 5.1.2.6
∂x j ∂x j ∂x j ∂x j ∂x j ∂x j ∂x j ∂x j
F.E Method is used only for deviatoric strain rate calculation in A.L.E and Euler formulation.
Volumetric strain rate is computed separately by transport of density and volume variation.
5.1.3 Assumed strain rate

Using Voigt convention, the strain rate of EQ. 5.1.2.5 can be written as:
8
{εɺ} = [B]{v} = ∑[BI ]{vI } EQ. 5.1.3.1
I =1
01-Jan-2017 7
With,
{εɺ} = εɺxx
t
εɺyy εɺzz 2εɺxy 2εɺyz 2εɺxz
t
∂ΦI ∂ΦI ∂ΦI 
 0 0 0 
 ∂x ∂x ∂y 
[BI ] =  0 ∂ΦI 0 ∂ΦI 0 ∂ΦI 
∂y ∂x ∂z
 ∂ΦI ∂ΦI ∂ΦI 
 0 0 0 
 ∂z ∂y ∂z 
It is useful to take the Belytschko-Bachrach's mix form [27] of the shape functions written by:
4
Φ I ( x, y, z, ξ ,η , ζ ) = ∆ I + bxI ⋅ x + byI ⋅ y + bzI ⋅ z + ∑ γ Iα φα EQ. 5.1.3.2
α =1
Where,
∂Φ I
biI = (ξ = η = ζ = 0);
∂xi
1  8
  8
  8
 
γ Iα = ΓIα −  ∑ ΓJα x J bxI −  ∑ ΓJα y J b yI −  ∑ ΓJα z J bzI ;
8  J =1   J =1   J =1  
φ = ηζ ξζ ξη ξηζ
The derivation of the shape functions is given by:
∂Φ I 4
= biI + ∑ γ Iα
∂φα
∂xi EQ. 5.1.3.3
∂xi α =1
It is decomposed by a constant part which is directly formulated with the Cartesian coordinates, and a non-
constant part which is to be approached separately. For the strain rate, only the non-constant part is modified by
the assumed strain. You can see in the following that the non-constant part or the high order part is just the
hourglass terms.
You now have the decomposition of the strain rate:
{εɺ} = ∑[BI ]{vI } = ∑([BI ]0 +[BI ]H ){vI } = {εɺ}0 +{εɺ}H

8 8
EQ. 5.1.3.4
I =1 I =1
with:
bxI 0 0  t
 4 α ∂φα 4 4

∑γI ∂x ∑ ∑
0 b  α ∂φα α ∂φα
0 0 0 0 γ I ∂x γI ∂y 
 yxI
 0 0 b  α=1 4 4
α
=1 α=1
4 
[BI ]0 =  zI
 ; [BI ] =  0
H
 ∑ γα φ ∂α
I ∂y 0 ∑ γα φ ∂α
I ∂x 0 ∑ α ∂φα 
γI ∂z

b b
 yxI xI 0  α
=1 α=1 α=1
 bzI 0 bxI   4 4 4 
∑γIα ∂zα ∑γIα ∂yα ∑
∂φ ∂φ ∂φα
   0 0 γα I ∂z 0 
 0 b byxI
 α=1 α=1 α
=1 
zI 
01-Jan-2017 8
Belvtschko and Bindeman [65] ASQBI assumed strain is used:
{εɺ} = ∑([BI ]0 +[BI ]H ){vI }

8
EQ. 5.1.3.5
I =1
with
 X1234I −νYI3 −νYI24 −νZI2 −νZI34

 
−ν XI −νXI −νZ1I −νZI34
3 14
YI1234
−ν X2 −νX14 −νY1 −νY24 
[ ]
H
BI =  I
12
I I
12
I Z1234
I

 YI XI 0 
 Z13 0 XI 13 
 I

 0 ZI23 YI23 
∂φ1 ∂φ3 ∂φ1 ∂φ3

where X I13 = γ I1 ∂x + γ I3 ∂x ; YI13 = γ 1I ∂y + γ I3 ∂y ;ν = 1ν−ν ;
To avoid shear locking, some hourglass modes are eliminated in the terms associated with shear so that no shear
strain occurs during pure bending. E.g.: YI3 , X I3 in εɺxy terms and all fourth hourglass modes in shear terms are
also removed since this mode is non-physical and is stabilized by other terms in [B ]
I
H
.
The terms with Poisson coefficient are added to obtain an isochoric assumed strain field when the nodal velocity
is equivoluminal. This avoids volumetric locking as ν = 0.5 . In addition, these terms enable the element to
capture transverse strains which occurs in a beam or plate in bending. The plane strain expressions are used since
this prevents incompatibility of the velocity associated with the assumed strains.
5.1.3.1 Incompressible or quasi-incompressible cases

(Flag for new solid element: Icpr =0,1,2)
For incompressible or quasi- incompressible materials, the new solid elements have no volume locking problem
due to the assumed strain. Another way to deal with this problem is to decompose the stress field into the
spherical part and the deviatory part and use reduced integration for spherical part so that the pressure is
constant. This method has the advantage on the computation time, especially for the full integrated element. For
some materials which the incompressibility can be changed during computation (for example: elastoplastic
material, which becomes incompressible as the growth of plasticity), the treatment is more complicated. Since
the elastoplastic material with large strain is the most frequently used, the constant pressure method has been
chosen for RADIOSS usual solid elements. The flag Icpr has been introduced for new solid elements.
• Icpr =0: assumed strain with ν terms is used.
• Icpr =1: assumed strain without ν terms and with a constant pressure method is used. The method is
recommended for incompressible (initial) materials.
• Icpr =2: assumed strain with ν terms is used, where ν is variable in function of the plasticity state.
The formulation is recommended for elastoplastic materials.
5.1.4 Internal force calculation

Internal forces are computed using the generalized relation:
∂Φ I
f iIint = ∫ σ ij dΩ EQ. 5.1.4.1
Ω
∂x j
01-Jan-2017 9
However, to increase the computational speed of the process, some simplifications are applied.
5.1.4.1 Reduced Integration Method

This is the default method for computing internal forces. A one point integration scheme with constant stress in
the element is used. Due to the nature of the shape functions, the amount of computation can be substantially
reduced:
∂Φ1 ∂Φ 7 ∂Φ 2 ∂Φ 8 ∂Φ 3 ∂Φ 5 ∂Φ 4 ∂Φ 6
=− ; =− ; =− ; =− EQ. 5.1.4.2
∂x j ∂x j ∂x j ∂x j ∂x j ∂x j ∂x j ∂x j
∂Φ I
Hence, the value is taken at the integration point and the internal force is computed using the relation:
∂x j
 ∂Φ I 
FiI = σ ij   Ω EQ. 5.1.4.3
 ∂x j 
 0
The force calculation is exact for the special case of the element being a parallelepiped.
5.1.4.2 Full Integration Method

The final approach that can be used is the full generalized formulation found in EQ. 5.1.4.1. A classical eight
point integration scheme, with non-constant stress, but constant pressure is used to avoid locking problems. This
is computationally expensive, having eight deviatoric stress tensors, but will produce accurate results with no
hourglass.
When assumed strains are used with full integration (HA8 element), the reduced integration of pressure is no
more necessary, as the assumed strain is then a free locking problem.
5.1.4.3 Improved Integration Method for ALE

This is an ALE method for computing internal forces (flag INTEG). A constant stress in the element is used.
∂Φ I
The value ∫ ∂x
Ω j
dΩ is computed with Gauss points.
5.1.5 Hourglass modes

Hourglass modes are element distortions that have zero strain energy. Thus, no stresses are created within the
element. There are 12 hourglass modes for a brick element, 4 modes for each of the 3 coordinate directions. Γ
represents the hourglass mode vector, as defined by Flanagan-Belytschko [12]. They produce linear strain
modes, which cannot be accounted for using a standard one point integration scheme.
Γ1 = (+ 1,−1,+1,−1,+1,−1,+1,−1)
01-Jan-2017 10
Γ 2 = (+ 1,+1,−1,−1,−1,−1,+1,+1)
Γ 3 = (+ 1,−1,−1,+1,−1,+1,+1,−1)
Γ 4 = (+ 1,−1,+1,−1,−1,+1,−1,+1)
To correct this phenomenon, it is necessary to introduce anti-hourglass forces and moments. Two possible
formulations are presented hereafter.
5.1.5.1 Kosloff & Frasier Formulation [10]

The Kosloff-Frasier hourglass formulation uses a simplified hourglass vector. The hourglass velocity rates are
defined as:
∂qiα 8
= ∑ ΓIα ⋅ viI EQ. 5.1.5.1
∂t I =1
where:
• Γ is the non-orthogonal hourglass mode shape vector,
• ν is the node velocity vector,
• i is the direction index, running from 1 to 3,
• I is the node index, from 1 to 8,
• α is the hourglass mode index, from 1 to 4.
This vector is not perfectly orthogonal to the rigid body and deformation modes.
All hourglass formulations except the physical stabilization formulation for solid elements in RADIOSS use a
viscous damping technique. This allows the hourglass resisting forces to be given by:
( ) ∑ ∂∂qt
α
1 2
f iIhgr = ρch 3 Ω i
⋅ ΓIα EQ. 5.1.5.2
4 α
01-Jan-2017 11
where:
• ρ is the material density,
• c is the sound speed,

• h is a dimensional scaling coefficient defined in the input,
• Ω is the volume.
5.1.5.2 Flanagan-Belytschko Formulation [12]

α
In the Kosloff-Frasier formulation seen in section 5.1.5.1, the hourglass base vector ΓI is not perfectly
orthogonal to the rigid body and deformation modes that are taken into account by the one point integration
scheme. The mean stress/strain formulation of a one point integration scheme only considers a fully linear
velocity field, so that the physical element modes generally contribute to the hourglass energy. To avoid this, the
idea in the Flanagan-Belytschko formulation is to build an hourglass velocity field which always remains
orthogonal to the physical element modes. This can be written as:
viIHour = viI − viILin EQ. 5.1.5.3
The linear portion of the velocity field can be expanded to give:
 ∂v
(
viIHour = viI −  viI + iI ⋅ x j − x j
 ∂x j
) EQ. 5.1.5.4
 
Decomposition on the hourglass vectors base gives [12]:
∂qiα  ∂v 
= ΓIα ⋅ viIHour =  viI − il ⋅ x j  ⋅ ΓIα EQ. 5.1.5.5
∂t  ∂x j 
 
where:
∂qiα
are the hourglass modal velocities,
∂t
ΓIα is the hourglass vectors base.
∂vi ∂Φ j
Remembering that = ⋅ viJ and factorizing EQ. 5.1.5.5 gives:
∂x j ∂x j
∂qiα  ∂Φ j 
= viI ⋅  ΓIα − x j ΓIα  EQ. 5.1.5.6
∂t  ∂x j 
 
∂Φ j
γ Iα = ΓIα − x j ΓJα EQ. 5.1.5.7
∂x j
is the hourglass shape vector used in place of ΓIα in EQ. 5.1.5.2.
01-Jan-2017 12
5.1.5.3 physical hourglass formulation HEPH

You also try to decompose the internal force vector as follows:
{f } = {( f ) }+ {( f )
I
int
I
int 0
I
int H
} EQ. 5.1.5.8
In elastic case, you have:
{f } = ∫ [B ] [C ]∑ [B ]{v }dΩ t 8
int J
I I J
Ω
j =1
EQ. 5.1.5.9
(
= ∫ [B I ] + [B I ] ) [C ]∑ ([B ] + [B ] ){v }dΩ
8
0 H t 0 H J
J J
Ω j =1
The constant part {( f ) }= ∫ ([B ] ) [C ]∑ [B ] {v }dΩ is evaluated at the quadrature point just like
I
int 0
I
0 t
8
j =1
J
0 J
Ω
other one-point integration formulations mentioned before, and the non-constant part (Hourglass) will be
calculated as following:
∂xi
Taking the simplification of = 0; (i ≠ j ) (that is the Jacobian matrix of EQ. 5.1.2.1 is diagonal), you
∂ξ j
have:
(f )
4
int H
iI = ∑ Qiα γ Iα EQ. 5.1.5.10
α =1
.
with 12 generalized hourglass stress rates Q iα calculated by:
[
Q ii = µ (H jj + H kk )qɺ ii + H ij qɺ jj + H ik qɺ kk ]
.
.
 1 
Q jj = µ  H ii qɺ ij + ν H ij qɺ ij  EQ. 5.1.5.11
1 − ν 
. 1 +ν
Q i 4 = 2µ H ii qɺ 4j
3
and
 ∂φ j 
2 2 2
 ∂φ   ∂φ 
H ii = ∫   dΩ = ∫  k  dΩ = 3∫  4  dΩ
Ω
∂xi  Ω
∂xi  Ω
∂xi  EQ. 5.1.5.12
∂φ ∂φ j
H ij = ∫ i dΩ
Ω
∂x j ∂xi
α
Where i,j,k are permuted between 1 to 3 and qɺ i has the same definition than in EQ. 5.1.5.6.
Extension to non-linear materials has been done simply by replacing shear modulus µ by its effective tangent
values which is evaluated at the quadrature point. For the usual elastoplastic materials, use a more sophistic
procedure which is described in the following section.
01-Jan-2017 13
5.1.5.3.1 Advanced elasto-plastic hourglass control

With one-point integration formulation, if the non-constant part follows exactly the state of constant part for the
case of elasto-plastic calculation, the plasticity will be under-estimated due to the fact that the constant
equivalent stress is often the smallest one in the element and element will be stiffer. Therefore, defining a yield
criterion for the non-constant part seems to be a good idea to overcome this drawback.
Plastic yield criterion:
The von Mises type of criterion is written by:
f = σ eq2 (ξ ,η , ζ ) − σ y2 = 0 EQ. 5.1.5.13
for any point in the solid element, where σy is evaluated at the quadrature point.
As only one criterion is used for the non-constant part, two choices are possible:
1. taking the mean value, i.e.: f ( )

= f σ eq ; σ eq =
1
Ω Ω∫
σ eq dΩ
2. taking the value by some representative points, for example: eight Gausse points
The second choice has been used in this element.
Elastro-plastic hourglass stress calculation:
The incremental hourglass stress is computed by:
• Elastic increment
(σ i )trH = (σ i )n + [C ]{εɺ} ∆t
H H
n+1
• Check the yield criterion

• If f ≥ 0 , the hourglass stress correction will be done by un radial return
(σ i )nH+1 = P((σ i )trH

n +1
,f)
5.1.6 Stability
The stability of the numerical algorithm depends on the size of the time step used for time integration (section
4.5). For brick elements, RADIOSS uses the following equation to calculate the size of the time step:
l
h≤k
(
c α + α 2 +1 ) EQ. 5.1.6.1
This is the same form as the Courant condition for damped materials. The characteristic length of a particular
element is computed using:
Element Volume
l= EQ. 5.1.6.2
Largest Side Surface
For a 6-sided brick, this length is equal to the smallest distance between two opposite faces.
The terms inside the parentheses in the denominator are specific values for the damping of the material:
2v
• α=
ρcl
• ν effective kinematic viscosity,
01-Jan-2017 14
1∂p
• c= for fluid materials,
ρ∂ρ
K 4µ λ + 2µ
• c= + = for a solid elastic material,
ρ 3ρ ρ
• K is the bulk modulus,
• λ ,µ are Lame moduli.
The scaling factor k=0.90, is used to prevent strange results that may occur when the time step is equal to the
Courant condition. This value can be altered by the user.
5.1.7 Shock waves

Shocks are non-isentropic phenomena, i.e. entropy is not conserved, and necessitates a special formulation.
The missing energy is generated by an artificial bulk viscosity q as derived by von Neumann and Richtmeyer [9].
This value is added to the pressure and is computed by:
 ∂ε  ∂ε
2
q = qa ρl  kk  − qb ρlc kk
2 2
EQ. 5.1.7.1
 ∂t  ∂t
where
• l is equal to 3
Ω or to the characteristic length,
• Ω is the volume,
∂ε kk
• is the volumetric compression strain rate tensor,
∂t
• c is the speed of sound in the medium.
The values of qa and qb are adimensional scalar factors defined as:
• qa is a scalar factor on the quadratic viscosity to be adjusted so that the Hugoniot equations are
verified. This value is defined by the user. The default value is 1.10.
• qb is a scalar factor on the linear viscosity that damps out the oscillations behind the shock. This is
user specified. The default value is 0.05.
Default values are adapted for velocities lower than Mach 2. However for viscoelastic materials (law 34, 35, 38)
or honeycomb (law 28), very small values are recommended, i.e. 10-20.
5.1.8 Element degeneration

Element degeneration is the collapsing of an element by one or more edges. For example: making an eight node
element into a seven node element by giving nodes 7 and 8 the same node number.
The use of degenerated elements for fluid applications is not recommended. The use of degenerated elements for
assumed strain formulation is not recommended. If they cannot be avoided, any two nodes belonging to a same
edge can be collapsed, with some examples shown below.
For solid elements, it is recommended that element symmetry be maintained.
For 4 node elements, it is recommended that the special tetrahedron element be used.
01-Jan-2017 15
Some examples of element degeneration are shown below.
Not recommended degenerations
Recommended degeneration
Connectivity: 1 2 3 4 5 5 5 5
Connectivity: 1 2 3 4 5 6 6 5
01-Jan-2017 16
5.1.9 Internal stress calculation

5.1.9.1 Global formulation
The time integration of stresses has been stated earlier (section 2.6.) as:
σ ij (t + ∆t ) = σ ij (t ) + σɺ ij ∆t EQ. 5.1.9.1
The stress rate is comprised of two components:
σɺ ij = σɺ ijv + σɺ ijr EQ. 5.1.9.2
where
• σɺ ijr is the stress rate due to the rigid body rotational velocity,
• σɺ ijv is the Jaumann objective stress tensor derivative.
The correction for stress rotation from time t to time t+ ∆t is given by [2]:
σɺ ijr = σ ik Ω kj + σ jk Ω ki EQ. 5.1.9.3
where Ω is the rigid body rotational velocity tensor (EQ. 2.4.1.11).

The Jaumann objective stress tensor derivative σɺ ijv is the corrected true stress rate tensor without rotational
effects. The constitutive law is directly applied to the Jaumann stress rate tensor.
Deviatoric stresses and pressure (see section 2.7) are computed separately and related by:
σ ij = sij − pδ ij EQ. 5.1.9.4
where
• sij is the deviatoric stress tensor;

• p is the pressure or mean stress - defined as positive in compression,
• δ ij is the substitution tensor or unit matrix.
5.1.9.2 Co-rotational Formulation

A co-rotational formulation for bricks is a formulation where rigid body rotations are directly computed from the
element's node positions. Objective stress and strain tensors are computed in the local (co-rotational) frame.
Internal forces are computed in the local frame and then rotated to the global system.
So, when co-rotational formulation is used, EQ. 5.1.10.2 σɺ ij = σɺ ijv + σɺ ijr reduces to:
σɺ ij = σɺ ijv EQ. 5.1.9.5
where σɺ ijv is the Jaumann objective stress tensor derivative expressed in the co-rotational frame.
The following illustrates orthogonalization, when one of the r, s, t directions is orthogonal to the two other
directions.
01-Jan-2017 17
Isoparametric frames Local (co-rotational)

When large rotations occur, this formulation is more accurate than the global formulation, for which the stress
rotation due to rigid body rotational velocity is computed in an incremental way.
Co-rotational formulation avoids this kind of problem.
Let us consider the following test:
Fix constant velocity on the top of the

The increment of the rigid body rotation vector during time step ∆t is:
(∂vx / ∂y − ∂v y / ∂x ) = 0

∆Ω = ∆t / 2 ⋅ (∂v x / ∂z − ∂v z / ∂x ) = ∂vx / ∂z EQ. 5.1.9.6
(∂v y / ∂x − ∂v x / ∂y ) = 0

So, ∆Ω y = α∆T / 2 where α = v / h equals the imposed velocity on the top of the brick divided by the
height of the brick (constant value).
Due to first order approximation, the increment of stress σ xx due to the rigid body motion is:
∆σ xxr = ∆Ω y (τ xz + τ zx ) = 2∆Ω yτ xz = α∆Tτ xz EQ. 5.1.9.7
Increment of stress σ zz due to the rigid body motion:
∆σ zzr = −∆Ω y (τ xz + τ zx ) = −2∆Ω yτ xz = −α∆Tτ xz EQ. 5.1.9.8
Increment of shear stress τ xz due to the rigid body motion:
∆τ xzr = ∆Ω y (σ zz − σ xx ) = 2∆Ω yσ zz = α∆Tσ zz EQ. 5.1.9.9
Increment of shear strain:
∆γ xz = ∆T (∂v x / ∂z + ∂v z / ∂x ) = α∆T EQ. 5.1.9.10
Increment of stress σ zz due to strain:
∆σ zzv = 0 EQ. 5.1.9.11
01-Jan-2017 18
and increment of shear stress due to strain is:
∆τ xzv = G∆γ xz = Gα∆T EQ. 5.1.9.12
where G is the shear modulus (material is linear elastic).

From EQ. 5.1.9.8 to EQ. 5.1.9.12, you have:
∆τ xz = α∆Tσ zz + Gα∆T 

∆σ = −α∆τ  EQ. 5.1.9.13
 zz xz 
System EQ. 5.1.9.13 leads to:
∆τ xz / ∆T 2 = −α 2τ xz EQ. 5.1.9.14
So, shear stress is sinusoidal and is not strictly increasing.
So, it is recommended to use co-rotational formulation, especially for visco-elastic materials such as foams, even
if this formulation is more time consuming than the global one.
5.1.9.3 Co-rotational formulation and orthotropic material

When orthotropic material and global formulation are used, the fiber is attached to the first direction of the
isoparametric frame and the fiber rotates a different way depending on the element numbering (see below).
01-Jan-2017 19
On the other hand, when the co-rotational formulation is used, the orthotropic frame keeps the same orientation
with respect to the local (co-rotating) frame, and is therefore also co-rotating (see below).
5.1.10 Deviatoric stress calculation

With the stress being separated into deviatoric and pressure (hydrostatic) stress (Section 2.7), it is the deviatoric
stress that is responsible for the plastic deformation of the material. The hydrostatic stress will either shrink or
expand the volume uniformly, i.e. with proportional change in shape. The determination of the deviatoric stress
tensor and whether the material will plastically deform requires a number of steps.
STEP 1: Perform an Elastic Calculation
The deviatoric stress is time integrated from the previous known value using the strain rate to compute an elastic
trial stress:
 
sijel (t + ∆t ) = sij (t ) + sɺijr ∆t + 2G  εɺij − εɺkkδ ij ∆t
1
EQ. 5.1.10.1
 3 
where G is the shear modulus.
This relationship is Hooke's Law, where the strain rate is multiplied by time to give strain.
STEP 2: Compute von Mises Equivalent Stress and Current Yield Stress
Depending on the type of material being modeled, the method by which yielding or failure is determined will
vary. The following explanation relates to an elastoplastic material (LAW2).
The von Mises equivalent stress relates a three dimensional state of stress back to a simple case of uniaxial
tension where material properties for yield and plasticity are well known and easily computed.
The von Mises stress, which is strain rate dependent, is calculated using the equation:
3 el el
σ vm
e
= sij sij EQ. 5.1.10.2
2
The flow stress is calculated from the previous plastic strain:
σ y (t ) = a + bε p (t )
n
EQ. 5.1.10.3
01-Jan-2017 20
For material types 3, 4, 10, 21, 22, 23 and 36, EQ. 5.1.11.3 is modified according to the different modeling of the
material curves.
STEP 3: Plasticity Check
The state of the deformation must be checked.
σ vm
e
−σ y ≤ 0
If this equation is satisfied, the state of stress is elastic. Otherwise, the flow stress has been exceeded and a
plasticity rule must be used. This is shown in Figure 5.1.2.
Figure 5.1.2 - Plasticity Check
The plasticity algorithm used is due to Mendelson, [1].

STEP 4: Compute Hardening Parameter
The hardening parameter is defined as the slope of the strain-hardening part of the stress-strain curve:
dσ y
H= EQ. 5.1.10.4
dε p
This is used to compute the plastic strain at time t:
σ vm − σ y
εɺ p ∆t = EQ. 5.1.10.5
3G + H
This plastic strain is time integrated to determine the plastic strain at time t + ∆t :
ε p (t + ∆t ) = ε p (t ) + εɺ p ∆t EQ. 5.1.10.6
The new flow stress is found using:
σ y (t + ∆t ) = a + bε p (t + ∆t )
n
EQ. 5.1.10.7
STEP 5: Radial Return
There are many possible methods for obtaining sijpa from the trial stress. The most popular method involves a
simple projection to the nearest point on the flow surface, which results in the radial return method.
The radial return calculation is given in EQ. 5.1.10.8. Figure 5.1.3 is a graphic representation of radial return.
σ y el
sijpa = sij EQ. 5.1.10.8
σ vm
01-Jan-2017 21
Figure 5.1.3 - Radial Return
01-Jan-2017 22
5.2 Solid Tetrahedron Elements
5.2.1 4 node solid tetrahedron

The RADIOSS solid tetrahedron element is a 4 node element with one integration point and a linear shape
function.
This element has no hourglass. But the drawbacks are the low convergence and the shear locking.
5.2.2 10 node solid tetrahedron

The RADIOSS solid tetrahedron element is a 10 nodes element with 4 integration points and a quadratic shape
function as shown in Figure 5.2.1.
Figure 5.2.1 – (a) Isoparametric 10 node tetrahedron , (b) Nodal mass distribution
Introducing volume coordinates in an isoparametric frame:

L1 = r
L2 = s
L3 = t
L4 = 1 − L1 − L2 − L3
01-Jan-2017 23
The shape functions are expressed by:
Φ1 = (2 L1 − 1)L1 EQ. 5.2.2.1
Φ 2 = (2 L2 − 1)L2 EQ. 5.2.2.2
Φ 3 = (2 L3 − 1)L3 EQ. 5.2.2.3
Φ 4 = (2 L4 − 1)L4 EQ. 5.2.2.4
Φ 5 = 4 L1 L2 EQ. 5.2.2.5
Φ 6 = 4 L2 L3 EQ. 5.2.2.6
Φ 7 = 4 L3 L1 EQ. 5.2.2.7
Φ 8 = 4 L1 L4 EQ. 5.2.2.8
Φ 9 = 4 L2 L4 EQ. 5.2.2.9
Φ10 = 4 L3 L4 EQ. 5.2.2.10
Location of the 4 integration points is expressed by [49].
With,
α = 0 ⋅ 58541020 and β = 0 ⋅13819660
a, b, c, and d are the 4 integration points.
5.2.2.1 Advantages and drawbacks

This element has various advantages:
• No hourglass
• Compatible with powerful mesh generators
• Fast convergence
• No shear locking.
But there are some drawbacks too:
• Low time step
• Not compatible with ALE formulation
• No direct compatibility with contact interface and other elements.
5.2.2.2 Time step

The time step for a regular tetrahedron is computed as:
01-Jan-2017 24
Lc
dt = EQ. 5.2.2.11
c
Where, Lc is the characteristic length of element depending on tetra type. The different types are shown in the
following figures:
For a regular 4 node tetra as shown in Figure 5.2.2:
2
Lc = a ; Lc = 0.816a EQ. 5.2.2.12
3
Figure 5.2.2 - 4 nodes tetra
For a regular 10 node tetra as shown in Figure 5.2.3:
5/ 2
Lc = a ; Lc = 0.264a EQ. 5.2.2.13
6
Figure 5.2.3 - 10 nodes tetra
For another regular tetra obtained by the assemblage of four hexa as shown in Figure 5.2.4, the characteristic
length is:
2/3
Lc = a ; Lc = 0.204a EQ. 5.2.2.14
4
Figure 5.2.4 - Other regular tetra
5.2.2.3 CPU cost: Time/Element/Cycle

The CPU cost is shown in Figure 5.2.5:
01-Jan-2017 25
Figure 5.2.5 - CPU cost in TEC
5.2.2.4 Comparison example

Below is a comparison of the 3 types of elements (8-nodes brick, 10-nodes tetra and 20-nodes brick). The results
are shown in Figure 5.2.6 for a plastic strain contour.
Figure 5.2.6 - Comparison (plastic strain max = 60%)
8 nodes brick 10 nodes tetras 20 nodes brick
01-Jan-2017 26
5.3 SHELL ELEMENTS

Since the degenerated continuum shell element formulation was introduced by Ahmad et al.[38], it has become
dominant in commercial Finite Element codes due to its advantage of being independent of any particular shell
theory, versatile and cost effective, and applicable in a reliable manner to both thin and thick shells.
In the standard 4-node shell element, full integration and reduced integration schemes have been used to
compute the stiffness matrices and force vectors:
- The full integration scheme is often used in programs for static or dynamic problems with implicit time
integration. It presents no problem for stability, but sometimes involves “locking” and computations are often
more expensive.
- The reduced integration scheme, especially with one-point quadrature (in the mid-surface), is widely used in
programs with explicit time integration such as RADIOSS and other programs applied essentially in
crashworthiness studies. These elements dramatically decrease the computation time, and are very competitive if
the hourglass modes (which result from the reduced integration scheme) are “well” stabilized.
5.3.1 Introduction
The historical shell element in RADIOSS is a simple bilinear Mindlin plate element coupled with a reduced
integration scheme using one integration point. It is applicable in a reliable manner to both thin and moderately
thick shells.
This element is very efficient if the spurious singular modes, called “hourglass modes”, which result from the
reduced integration are stabilized.
The stabilization approach consists of providing additional stiffness so that the spurious singular modes are
suppressed. Also, it offers the possibility of avoiding some locking problems. One of the first solutions was to
generalize the formulation of Kosloff and Frazier [10] for brick element to shell element. It can be shown that
the element produces accurate flexural response (thus, free from the membrane shear locking) and is equivalent
to the incompatible model element of Wilson et al. [21] without the static condensation procedure. Taylor [47]
extended this work to shell elements. Hughes and Liu [22] employed a similar approach and extended it to non-
linear problems.
Belytschko and Tsay [23] developed a stabilized flat element based on the γ projections developed by Flanagan
and Belytschko [12]. Its essential feature is that hourglass control is orthogonal to any linear field, thus
preserving consistency. The stabilized stiffness is approached by a diagonal matrix and scaled by the
perturbation parameters hi which are introduced as a regulator of the stiffness for nonlinear problems. The
parameters hi are generally chosen to be as small as possible, so this approach is often called, perturbation
stabilization.
The elements with perturbation stabilization have two major drawbacks:
• The parameters hi are user-inputs and are generally problem-dependent.

• Poor behavior with irregular geometries (in-plane, out-of-plane). The stabilized stiffness (or stabilized
forces) is often evaluated based on a regular flat geometry, so they generally do not pass either the
Patch-test or the Twisted beam test.
Belytschko et al. [17] extended this perturbation stabilization to the 4-node shell element which has become
widely used in explicit programs.
Belytschko et al. [24] improved the poor behavior exhibited in the warped configuration by adding a coupling
curvature-translation term, and a particular nodal projection for the transverse shear calculation analogous to that
developed by Hughes and Tezduyar [25], and MacNeal [26]. This element passes the Kirchhoff patch test and
the Twisted beam test, but it cannot be extended to a general 6 DOF element due to the particular projection.
Belytschko and Bachrach [27] used a new method called “physical stabilization” to overcome the first drawback
of the quadrilateral plane element. This method consists of explicitly evaluating the stabilized stiffness with the
help of 'assumed strains', so that no arbitrary parameters need to be prescribed. Engelmann and Whirley [28]
have applied it to the 4-node shell element. An alternative way to evaluate the stabilized stiffness explicitly is
given by Liu et al. [29] based on Hughes and Liu's 4-node selected reduced integration scheme element [22], in
01-Jan-2017 27
which the strain field is expressed explicitly in terms of natural coordinates by a Taylor-series expansion. A
remarkable improvement in the one-point quadrature shell element with physical stabilization has been
performed by Belytschko and Leviathan [18]. The element performs superbly for both flat and warped elements
especially in linear cases, even in comparison with a similar element under a full integration scheme, and is only
20% slower than the Belytschko and Tsay element. More recently, based on Belytschko and Leviathan's
element, Zhu and Zacharia [30] implemented the drilling rotation DOF in their one-point quadrature shell
element; the drilling rotation is independently interpolated by the Allman function [39] based on Hughes and
Brezzi's [41] mixed variational formulation.
The physical stabilization with assumed strain method seems to offer a rational way of developing a cost
effective shell element with a reduced integration scheme. The use of the assumed strains based on the mixed
variational principles, is powerful, not only in avoiding the locking problems (volumetric locking, membrane
shear locking, as in Belytschko and Bindeman [31]; transverse shear locking, as in Dvorkin and Bathe [32]), but
also in providing a new way to compute stiffness. However, as highlighted by Stolarski et al. [33], assumed
strain elements generally do not have rigorous foundations; there is almost no constraint for the independent
assumed strains interpolation. Therefore, a sound theoretical understanding and numerous tests are needed in
order to prove the legitimacy of the assumed strain elements.
The greatest uncertainty of the one-point quadrature shell elements with physical stabilization is with respect to
the nonlinear problems. All of these elements with physical stabilization mentioned above rely on the
assumptions that the spin and the material properties are constant within the element. The first assumption is
necessary to ensure the objectivity principle in geometrical nonlinear problems. The second was adapted in order
to extend the explicit evaluation of stabilized stiffness for elastic problems to the physical nonlinear problems. It
is found that the second assumption leads to a theoretical contradiction in the case of an elastoplastic problem (a
classic physical nonlinear problem), and results in poor behavior in case of certain crash computations.
Zeng and Combescure [15] have proposed an improved 4-node shell element named QPPS with one-point
quadrature based on the physical stabilization which is valid for the whole range of its applications (see the
Chapter 5.3.12). The formulation is based largely on that of Belytschko and Leviathan.
Based on the QPPS element, Zeng and Winkelmuller have developed a new improved element named QEPH
which is integrated in RADIOSS 44 version (see Chapter 5.3.13).
5.3.2 Bilinear Mindlin plate element

Most of the following explanation concerns four node plate elements, Figure 5.3.1. Section 5.3.13 explains the
three node plate element, shown in Figure 5.3.2.
Figure 5.3.1 - Four Node Plate Element
01-Jan-2017 28
Figure 5.3.2 - Three Node Plate Element
Plate theory assumes that one dimension (the thickness, z) of the structure is small compared to the other
dimensions. Hence, the 3D continuum theory is reduced to a 2D theory. Nodal unknowns are the velocities
( )
(v x′ v y′ vz ) of the mid plane and the nodal rotation rates ω x′ω y as a consequence of the suppressed z direction.
The thickness of elements can be kept constant, or allowed to be variable. This is user defined. The elements are
always in a state of plane stress, i.e. σ zz = 0 , or there is no stress acting perpendicular to the plane of the
element. A plane orthogonal to the mid-plane remains a plane, but not necessarily orthogonal as in Kirchhoff
∂v z ∂v z
theory, (where ε xz = ε yz = 0 ) leading to the rotations rates ωx = − and ω y = . In Mindlin plate
∂y ∂x
theory, the rotations are independent variables.
5.3.3 Time step for stability

The characteristic length, L, for computing the critical time step, referring back to Figure 5.3.3, is defined by:
area
L1 =
( )
max 13, 42
EQ. 5.3.3.1
L2 = min (12, 23, 34, 41,13, 42 ) EQ. 5.3.3.2
Lc = max(L1 , L2 ) EQ. 5.3.3.3
When the orthogonalized mode of the hourglass perturbation formulation is used, the characteristic length is
defined as:
L3 = max(L1 , L2 ) EQ. 5.3.3.4
L4 = 0.5
(L1 + L2 )
max (hm′ h f )
EQ. 5.3.3.5
Lc = min (L3 , L4 ) EQ. 5.3.3.6
where hm is the shell membrane hourglass coefficient and h f is the shell out of plane hourglass coefficient, as
mentioned in section 5.3.8.
5.3.4 Local reference frame

Three coordinate systems are introduced in the formulation:
• Global Cartesian fixed system (

X = Xi + Yj + Zk )
• Natural system (ξ ,η , ζ ) , covariant axes x,y
01-Jan-2017 29
• Local systems (x,y,z) defined by an orthogonal set of unit base vectors (t1,t2,n). n is taken to be
normal to the mid-surface coinciding with ζ , and (t1,t2 ) are taken in the tangent plane of the mid-
surface.
Figure 5.3.3 - Local Reference Frame
The vector normal to the plane of the element at the mid point is defined as:
x× y
n= EQ. 5.3.4.1
x× y
The vector defining the local direction is:
x y
t1 = t2 = EQ. 5.3.4.2
x y
Hence, the vector defining the local direction is found from the cross product of the two previous vectors:
t 2 = n × t1 EQ. 5.3.4.3
5.3.5 Bilinear shape functions

The shape functions defining the bilinear element used in the Mindlin plate are:
Φ I (ξ ,η ) = (1 + ξ I ξ )(1 + η Iη )
1
EQ. 5.3.5.1
4
or, in terms of local coordinates:
Φ I ( x, y ) = aI + bI x + cI y + d I xy EQ. 5.3.5.2
It is also useful to write the shape functions in the Belytschko-Bachrach mix form [27]:
Φ I ( x, y , ξη ) = ∆ I + bxI x + byI y + γ I ξη EQ. 5.3.5.3
01-Jan-2017 30
with
[ ]
∆ I = t I − (t I x I )bxI − (t I y I )byI ; t = (1,1,1,1)
bxI = ( y24 y31 y42 y13 ) / A ; (f = (f

ij i − f j )/ 2 )
byI = (x42 x13 x24 x31 ) / A
γ I = [ΓI − (ΓJ x J )bxI − (ΓJ y J )bxI ]/ 4 ; Γ = (1,−1,1,−1)

A is the area of the element
The velocity of the element at the mid-plane reference point is found using the relations:
4
vx = ∑ Φ I v xI EQ. 5.3.5.4
I =1
4
v y = ∑ Φ I v yI EQ. 5.3.5.5
I =1
4
vz = ∑ Φ I vzI EQ. 5.3.5.6
I =1
where, vxI , v yI , vzI are the nodal velocities in the x,y,z directions.
In a similar fashion, the element rotations are found by:
4
ω x = ∑ Φ I ω xI EQ. 5.3.5.7
I =1
4
ω y = ∑ Φ I ω yI EQ. 5.3.5.8
I =1
where ω xI and ω yI are the nodal rotational velocities about the x and y reference axes.
The velocity change with respect to the coordinate change is given by:
∂v x 4
∂Φ I
=∑ v xI EQ. 5.3.5.9
∂x I =1 ∂x
∂v x 4
∂Φ I
=∑ v xI EQ. 5.3.5.10
∂y I =1 ∂y
5.3.6 Mechanical properties

Shell elements behave in two ways, either membrane or bending behavior. The Mindlin plate elements that are
used by RADIOSS account for bending and transverse shear deformation. Hence, they can be used to model
thick and thin plates.
5.3.6.1 Membrane Behavior

The membrane strain rates for Mindlin plate elements are defined as:
∂v x
eɺxx = EQ. 5.3.6.1
∂x
01-Jan-2017 31
∂v y
eɺ yy = EQ. 5.3.6.2
∂y
1  ∂v ∂v y 
eɺxy =  x +  EQ. 5.3.6.3
2  ∂y ∂x 
1  ∂v ∂v  1  ∂v 
eɺxz =  x + z  =  ω y + z  EQ. 5.3.6.4
2  ∂z ∂x  2  ∂x 
1  ∂v y ∂v z  1 ∂v 
eɺ yz =  + =  − ω x + z  EQ. 5.3.6.5
2  ∂z ∂y  2 ∂y 
where eɺij is the membrane strain rate.
5.3.6.2 Bending Behavior

The bending behavior in plate elements is described using the amount of curvature. The curvature rates of the
Mindlin plate elements are defined as:
∂ω y
χɺ x = EQ. 5.3.6.6
∂x
∂ω x
χɺ y = − EQ. 5.3.6.7
∂y
1  ∂ω ∂ω 
χɺ xy =  y − x  EQ. 5.3.6.8
2  ∂y ∂x 
where χɺ ij is the curvature rate.
5.3.6.3 Strain Rate calculation

The calculation of the strain rate of an individual element is divided into two parts, membrane and bending strain
rates.
Membrane Strain rate
The vector defining the membrane strain rate is:
{eɺ}m = {eɺx′eɺ y′ 2eɺxy } EQ. 5.3.6.9
This vector is computed from the velocity field vector {v}m and the shape function gradient {B}m :
{eɺ}m = {B}m {v}m EQ. 5.3.6.10
where
{v}m = {v1x′v1y′vx2′v 2y′vx3′v 3y′vx4′v y4 } EQ. 5.3.6.11
 ∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
 0 0 0 0 
 ∂x ∂x ∂x ∂x 
∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
[B ]m = 0

0 0 0 EQ. 5.3.6.12
∂y ∂y ∂y ∂y 
 ∂Φ ∂Φ1 ∂Φ 2 ∂Φ 2 ∂Φ 3 ∂Φ 3 ∂Φ 4 ∂Φ 4 
 1 
 ∂y ∂x ∂y ∂x ∂y ∂x ∂y ∂x 
01-Jan-2017 32
Bending Strain rate

The vector defining the bending strain rate is:
{eɺ}b = {χɺ x′ χɺ y ′ 2 χɺ xy ′ 2eɺzx′ 2eɺ yz } EQ. 5.3.6.13
As with the membrane strain rate, the bending strain rate is computed from the velocity field vector. However,
the velocity field vector for the bending strain rate contains rotational velocities, as well as translations:
{eɺ}b = {B}b {v}b EQ. 5.3.6.14
where
{v}b = {ω1y ,−ω1x ,ω y2 ,−ω x2 ,ω y3 ,−ω x3 , ω y4 ,−ωx4 , v1z , vz2 , vz3 , vz4 } EQ. 5.3.6.15
 ∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
 ∂x 0 0 0 0 0 0 0 0 
∂x ∂x ∂x
 ∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
 0 0 0 0 0 0 0 0 
 ∂y ∂y ∂y ∂y 
∂Φ 3 ∂Φ 3
[B ]b =  ∂Φ1 ∂Φ1 ∂Φ 2 ∂Φ 2 ∂Φ 4 ∂Φ 4
 
0 0 0 0 
 ∂y ∂x ∂y ∂x ∂y ∂x ∂y ∂x 
Φ ∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
0 Φ2 0 Φ3 0 Φ4 0
 1 ∂x ∂x ∂x ∂x 
 ∂Φ1 ∂Φ 2 ∂Φ 3 ∂Φ 4 
 0 Φ1 0 Φ2 0 Φ3 0 Φ4 
 ∂y ∂y ∂y ∂y 
EQ. 5.3.6.16
Figure 5.3.4 - Strain rate calculation
01-Jan-2017 33
5.3.6.4 Mass and Inertia

Consider a rectangular plate with sides of length a and b, surface area A = ab and thickness t, as shown in Figure
5.3.5.
Figure 5.3.5 - Mass Distribution
z
4 3
y
b
x
1 2
m
a
Due to the lumped mass formulation used by RADIOSS, the lumped mass at a particular node is:
1
m= ρAt EQ. 5.3.6.17
4
The mass moments of inertia, with respect to local element reference frame, are calculated at node i by:
 b2 + t 2 
I xx = m  EQ. 5.3.6.18
 12 
 a2 + t 2 
I yy = m  EQ. 5.3.6.19
 12 
 a 2 + b2 
I zz = m  EQ. 5.3.6.20
 12 
ab
I xy = −m EQ. 5.3.6.21
16
5.3.6.5 Inertia Stability

With the exact formula for inertia (EQS 5.3.6.18 to 5.3.6.21), the solution tends to diverge for large rotation
rates. Belytschko proposed a way to stabilize the solution by setting I xx = I yy , i.e. to consider the rectangle as a
square with respect to the inertia calculation only. This introduces an error into the formulation. However, if the
aspect ratio is small the error will be minimal. In RADIOSS a better stabilization is obtained by:
 A t2 
I xx = m +  EQ. 5.3.6.22
 f 12 
I zz = I yy = I xx EQ. 5.3.6.23
I xy = 0 EQ. 5.3.6.24
where f is a regulator factor with default value f=12 for QBAT element and f=9 for other quadrilateral elements.
01-Jan-2017 34
5.3.7 Internal forces

The internal force vector is given by:
f int = ∫ B tσdΩ e EQ. 5.3.7.1

Ωe
In elasticity it becomes:
f int = ∫ B t CBvdΩ e EQ. 5.3.7.2

Ωe
It can be written as:

0
f int = f int + f int
hgr
EQ. 5.3.7.3
0
with the constant part f int being computed with one-point quadrature and the non constant part or hourglass
hgr
part f int being computed by perturbation stabilization (Ishell = 1, 2 ,3 ...) or by physical stabilization (Ishell =
22).
5.3.8 Hourglass modes

Hourglass modes are element distortions that have zero strain energy. The 4 node shell element has 12
translational modes, 3 rigid body modes (1, 2, 9), 6 deformation modes (3, 4, 5, 6, 10, 11) and 3 hourglass modes
(7, 8, 12).
Figure 5.3.6 - Translational Modes of Shell
Along with the translational modes, the 4 node shell has 12 rotational modes: 4 out of plane rotation modes (1, 2,
3, 4), 2 deformation modes (5, 6), 2 rigid body or deformation modes (7, 8) and 4 hourglass modes (9, 10, 11,
12).
01-Jan-2017 35
Figure 5.3.7 - Rotational Modes of Shell
5.3.8.1 Hourglass Viscous Forces

Hourglass resistance forces are usually either viscous or stiffness related. The viscous forces relate to the rate of
displacement or velocity of the elemental nodes, as if the material was a highly viscous fluid. The viscous
formulation used by RADIOSS is the same as that outlined by Kosloff and Frasier [10]. Refer to section 5.1.5.
An hourglass normalized vector is defined as:
Γ = (1, − 1, 1, − 1) EQ. 5.3.8.1

The hourglass velocity rate for the above vector is defined as:
∂qi
= ΓI viI = viI − vi 2 + vi 3 − vi 4 EQ. 5.3.8.2
∂t
The hourglass resisting forces at node I for in-plane modes are:
1 A ∂qi
f iIhgr = ρct hm ΓI EQ. 5.3.8.3
4 2 ∂t
For out of plane mode, the resisting forces are:
1 h ∂qi
f iIhgr = ρct 2 f ΓI EQ. 5.3.8.4
4 10 ∂t
where
i is the direction index,
I is the node index,
t is the element thickness,
c is the sound propagation speed,
A is the element area,
ρ is the material density,
01-Jan-2017 36
h m is the shell membrane hourglass coefficient,

h f is the shell out of plane hourglass coefficient.
5.3.8.2 Hourglass Elastic Stiffness Forces

RADIOSS can apply a stiffness force to resist hourglass modes. This acts in a similar fashion to the viscous
resistance, but uses the elastic material stiffness and node displacement to determine the size of the force. The
formulation is the same as that outlined by Flanagan et al. [12]. Refer to section 5.1.5.2. The hourglass resultant
forces are defined as:
f iIhgr = f i hgr ΓI EQ. 5.3.8.5
For membrane modes:
∂q
f i hgr (t + ∆t ) = f i hgr (t ) + hm Et i ∆t
1
EQ. 5.3.8.6
8 ∂t
For out of plane modes:
∂q
f i hgr (t + ∆t ) = f i hgr (t ) +
1
h f Et 3 i ∆t EQ. 5.3.8.7
40 ∂t
where
∆t is the time step,
E is Young's Modulus.
5.3.8.3 Hourglass Viscous Moments

This formulation is analogous to the hourglass viscous force scheme. The hourglass angular velocity rate is
defined for the main hourglass modes as:
∂ri
= ΓI ωiII = ωi1 − ωi 2 + ωi 3 − ωi 4 EQ. 5.3.8.8
∂t
The hourglass resisting moments at node I are given by:
1 hr ∂r
miIhgr = ρcAt 2 i ΓI EQ. 5.3.8.9
50 2 ∂t
where h r is the shell rotation hourglass coefficient.
5.3.9 Hourglass resistance

To correct this phenomenon, it is necessary to introduce anti-hourglass forces and moments. Two possible
formulations are presented hereafter.
5.3.9.1 Flanagan-Belytschko Formulation [12]

Ishell=1
α
In the Kosloff-Frasier formulation seen in section 5.1.5.1, the hourglass base vector ΓI is not perfectly
orthogonal to the rigid body and deformation modes that are taken into account by the one point integration
scheme. The mean stress/strain formulation of a one point integration scheme only considers a fully linear
velocity field, so that the physical element modes generally contribute to the hourglass energy. To avoid this, the
idea in the Flanagan-Belytschko formulation is to build an hourglass velocity field which always remains
orthogonal to the physical element modes. This can be written as:
viIHour = viI − viILin EQ. 5.3.9.1
01-Jan-2017 37
The linear portion of the velocity field can be expanded to give:
 ∂v
(
viIHour = viI −  viI + iI ⋅ x j − x j
 ∂x j
) EQ. 5.3.9.2
 
Decomposition on the hourglass base vectors gives [12]:
∂qiα  ∂v 
= ΓIα ⋅ viIHour =  viI − iI ⋅ x j  ⋅ ΓIα EQ. 5.3.9.3
∂t  ∂x j 
 
where
∂qiα
are the hourglass modal velocities,
∂t
ΓIα are the hourglass vectors, base.
∂vi ∂Φ j
Remembering that = ⋅ viJ and factorizing EQ. 5.1.5.5 gives:
∂x j ∂x j
∂qiα  ∂Φ J 
= viI ⋅  ΓIα − x j ΓJα  EQ. 5.3.9.4
∂t  ∂x j 
 
∂Φ J
γ Iα = ΓIα − x j ΓJα EQ. 5.3.9.5
∂x j
is the hourglass shape vector used in place of ΓIα in EQ. 5.1.5.2.

Figure 5.3.8 - Flanagan Belytschko Hourglass formulation
01-Jan-2017 38
5.3.9.2 Elastoplastic Hourglass Forces

Ishell=3
The same formulation as elastic hourglass forces is used (section 5.3.8.2 and Flanagan et al. [12]) but the forces
are bounded with a maximum force depending on the current element mean yield stress. The hourglass forces are
defined as:
f iIhgr = f i hgr ΓI EQ. 5.3.9.6
For in plane mode:
∂q
f i hgr (t + ∆t ) = f i hgr (t ) + hm Et i ∆t
1
EQ. 5.3.9.7
8 ∂t
 
f i hgr (t + ∆t ) = min  f i hgr (t + ∆t ),
1
hmσ yt A  EQ. 5.3.9.8
 2 
For out of plane mode:
∂q
f i hgr (t + ∆t ) = f i hgr (t ) +
1
h f Et 3 i ∆t EQ. 5.3.9.9
40 ∂t
 
f i hgr (t + ∆t ) = min f i hgr (t + ∆t ), h f σ y t 2 
1
EQ. 5.3.9.10
 4 
where:
σy is the yield stress,
A is the element area.
5.3.9.3 Physical Hourglass Forces

Ishell=22, 24
The hourglass forces are given by:
hgr
f int = ∫ B Ht CB H vdΩ e EQ. 5.3.9.11
Ωe
For in-plane membrane rate-of-deformation, with Φ = ξη and γ I defined in EQ 5.3.5.3:
γ I φ , x 0 0
[(B ) ]
m H
I

= 0 γ I φ , y 0 EQ. 5.3.9.12
γ I φ , y γ I φ , x 0
For bending:
 0 γ Iφ , x 
[(B ) ]
b H
I

= − γ Iφ , y 0  EQ. 5.3.9.13
 − γ I φ , x γ I φ , y 
It is shown in [16] that the non-constant part of the membrane strain rate does not vanish when a warped element
undergoes a rigid body rotation. Thus, a modified matrix [ (B )
m H
I ] is chosen using zγ = γ I z I as a measure of
the warping:
01-Jan-2017 39
γ Iφ, x 0 zγ bxI φ , x 
[(B ) ]
m H
I

= 0 γ Iφ, y zγ byIφ , y

 EQ. 5.3.9.14
γ I φ, y
 γ Iφ, x zγ (bxIφ , y +byIφ , x )
This matrix is different from the Belytschko-Leviathan [17] correction term added at rotational positions, which
couples translations to curvatures as follows:
 1 
γ Iφ, x 0 0 0 −
4
zγ φ, x 
[(B ) ]
m H
I

= 0

γ Iφ, y 0
1
4
zγ φ, y 0 


EQ. 5.3.9.15
γ φ 1 1
γ Iφ, x 0 zγ φ, x − zγ φ, y 
 I , y 4 4 
This will lead to “membrane locking” (the membrane strain will not vanish under a constant bending loading).
According to the general formulation, the coupling is presented in terms of bending and not in terms of
m
membrane, yet the normal translation components in ( BI ) do not vanish for a warped element due to the
tangent vectors ti (ξ ,η ) which differ from ti(0,0).
5.3.9.4 Full integrated formulation

Ishell=12
The element is based on the Q4γ24 shell element developed in [40] by Batoz and Dhatt. The element has 4 nodes
with 5 local degrees-of-freedom per node. Its formulation is based on the Cartesian shell approach where the
middle surface is curved. The shell surface is fully integrated with four Gauss points. Due to an in-plane reduced
integration for shear, the element shear locking problems are avoided. The element without hourglass
deformations is based on Mindlin-Reissner plate theory where the transversal shear deformation is taken into
account in the expression of the internal energy. The reader is invited to consult the reference for more details.
5.3.9.5 Shell membrane damping

The shell membrane damping, dm, is only used for law 25, 27, 19, 32 and 36. The Shell membrane damping
factor is a factor on the numerical VISCOSITY and not a physical viscosity. Its effect is shown in the formula of
the calculation of forces in a shell element:
dm = dm read in D00 input (Shell membrane damping factor parameter) then:
dm = 2 ⋅ dmD 00 ⋅ ρ 0 ⋅ c ⋅ AREA EQ. 5.3.9.16
Effect in the force vector (F) calculation:
 εɺ 
F1new = F1old + dm εɺ11 + 22  EQ. 5.3.9.17
 2 
 εɺ 
F2 new = F2 old + dm εɺ22 + 11  EQ. 5.3.9.18
 2 
εɺ12
F3new = F3old + dm EQ. 5.3.9.19
3
Where: ρ0 is the density
AREA is the area of the shell element surface

dt is the time step
c is the sound speed
01-Jan-2017 40
In order to calibrate the dm value so that it represents the physical viscosity, one should obtain the same size for
all shell elements (Cf. AREA factor), then scale the physical viscosity value to the element size.
5.3.10 Stress and strain calculation

The stress and strain for a shell element can be written in vector notation. Each component is a stress or strain
feature of the element. The generalized strain ε can be written as:
{ε } = {ex , e y , exy , k x , k y , k xy } EQ. 5.3.10.1
where
eij is the membrane strain,
χ ij is the bending strain or curvature.
The generalized stress Σ can be written as:
{∑} = {N x , N y , N xy , M x , M y , M xy } EQ. 5.3.10.2
t/2 t/2
where: Nx = ∫ σ x dz
−t / 2
M x = − ∫ σ x zdz
−t / 2
t/2 t/2
Ny = ∫ σ y dz
−t / 2
M y = − ∫ σ y zdz
−t / 2
t/2 t/2
N xy = ∫ σ xy dz
−t / 2
M xy = − ∫ σ xy zdz
−t / 2
t/2 t/2
N yz = ∫ σ yz dz
−t / 2
N xz = ∫σ
−t / 2
xz dz
5.3.10.1 Isotropic Linear Elastic Stress Calculation

The stress for an isotropic linear elastic shell for each time increment is computed using:
{∑ el
}
(t + ∆t ) = {∑(t )}+ L{εɺ}∆t EQ. 5.3.10.3
where EQ. 5.3.10.4
L 0
L= m
Lb 
EQ. 5.3.10.5
0
 Et − vEt 
1 − v 2 0 
1− v2
 − vEt Et 
Lm =  0  EQ. 5.3.10.6
1 − v 1− v2
2

 0 0
Et 
 1 + v 
01-Jan-2017 41
 Et 3 − vEt 3 

(
12 1 − v3
2
) (
12 1 − v 2 ) 0 

− vEt Et 3
Lb =  0 

(
12 1 − v 2 ) (
12 1 − v 2 ) 
EQ. 5.3.10.7
 Et 3 
0 0
 12(1 + v )
E is the Young's or Elastic Modulus,
ν is Poisson's Ratio,
t is the shell thickness.
5.3.10.2 Isotropic Linear Elastic-Plastic Stress Calculation

An incremental step-by-step method is usually used to resolve the nonlinear problems due to elasto-plastic
material behavior. The problem is presented by the resolution of the following equation:
σɺ = C : (εɺ − εɺ p ) EQ. 5.3.10.8
f (σ ,σ y ) = σ eq2 − σ y2 = 0 ; σɺ y = Hεɺ p EQ. 5.3.10.9
fɺ = 0 EQ. 5.3.10.10
∂f ɺ
and εɺ p = λ EQ. 5.3.10.11
∂σ
f is the yield surface function for plasticity for associative hardening. The equivalent stress σ eq may be
expressed in form:
σ eq2 = {σ }t [A]{σ } EQ. 5.3.10.12
σ xx  1 − 12 0
  
with {σ } = σ yy  and [ A] = − 12
 1 0 for von Mises criteria.
σ   0 0 3
 xy
The normality law (EQ. 5.3.10.11) for associated plasticity is written as:
∂f ɺ εɺ p
{εɺ } = λ = 2[ A]{σ }λɺ = [A]{σ } EQ. 5.3.10.13
∂ (σ ) σy
p
Where εɺ is the equivalent plastic deformation.

p
EQ. 5.3.10.8 is written in an incremental form:
{σ }n +1 = {σ }n + {dσ } = {σ }n + [C ]({dε }− {dε p }) = {σ ∗ }− dεn +1 [C ][A]{σ }n +1

p
EQ. 5.3.10.14
σy
01-Jan-2017 42
{ }
Where σ represents stress components obtained by an elastic increment and [C] the elastic matrix in plane
*
stress. The equations EQ. 5.3.10.8 to 14 lead to obtain the nonlinear equation:
( )
f dε p = 0 EQ. 5.3.10.15
that can be resolved by an iterative algorithm as Newton-Raphson method.
To determine the elastic-plastic state of a shell element, a number of steps have to be performed to check for
yielding and defining a plasticity relationship. Stress-strain and force-displacement curves for a particular ductile
material are shown in Figure 5.3.9.
Figure 5.3.9 - Material Curve
The steps involved in the stress calculation are as follows.

1. Strain calculation at integration point z
The overall strain on an element due to both membrane and bending forces is:
ε x = ex − zχ x EQ. 5.3.10.16
ε y = e y − zχ y EQ. 5.3.10.17
ε xy = exy − zχ xy EQ. 5.3.10.18
{ε } = {ε x , ε y , ε xy } EQ. 5.3.10.19
2. Elastic stress calculation

The stress is defined as:
{σ } = {σ x ,σ y ,σ xy } EQ. 5.3.10.20
01-Jan-2017 43
It is calculated using explicit time integration and the strain rate:
{σ (t + ∆t )} = {σ (t )} + L{εɺ}∆t
el
EQ. 5.3.10.21
 E vE 
1 − v 2 0 
1 − v2
 vE E 
L= 0 
1 − v 1 − v2
2

 0 0
E 
 1 + v 
The two shear stresses acting across the thickness of the element are calculated by:
σ yzel (t + ∆t ) = σ yz (t ) + α
E
eɺ yz ∆t EQ. 5.3.10.22
1+ v
σ xzel (t + ∆t ) = σ xz (t ) + α
E
eɺxz ∆t
1+ v
where α is the shear factor. Default is Reissner's value of 5/6.
3. von Mises yield criterion
The von Mises yield criterion for shell elements is defined as:
σ vm
2
= σ x2 + σ y2 − σ x σ y + 3σ xy2 EQ. 5.3.10.23
For type 2 simple elastic-plastic material, the yield stress is calculated using:
σ yield (t ) = a + bε p (t )
n
EQ. 5.3.10.24
This equation will vary according to the type of material being modeled.
4. Plasticity Check
The element's state of stress must be checked to see if it has yielded. These values are compared with the von
Mises and Yield stresses calculated in the previous step. If the von Mises stress is greater than the yield stress,
then the material will be said to be in the plastic range of the stress-strain curve.
Figure 5.3.10 - Plasticity Check
5. Compute plastically admissible stresses

If the state of stress of the element is in the plastic region, there are two different analyses that can be used as
described in the next paragraph. The scheme used is defined in the shell property set, card 2 of the input.
01-Jan-2017 44
6. Compute thickness change

The necking of the shells undergoing large strains in hardening phase can be taken into account by computing
normal strain ε zz in an incremental process. The incompressibility hypothesis in plasticity gives:
d ε zzp = − ( d ε xxp + d ε yyp ) EQ. 5.3.10.25
where the components of membrane strain dε xxp and dε yyp are computed by EQ. 5.3.10.13 as:
 dε xxp  dε p σ 
 p= [A]2×2  xx 
dε yy  σ y σ yy  EQ. 5.3.10.26
The plan stress condition dσ zz = 0 allows to resolve for dε zz :
υ 1 − 2υ
dε zz = − (dε + dε yy ) + dε zzp EQ. 5.3.10.27
1 −υ 1−υ
xx
5.3.10.3 Plastically Admissible Stresses

Radial return
Iplas=2
When the shell plane stress plasticity flag is set to 0 on card 1 of the shell property type definition, a radial return
plasticity analysis is performed. Thus, Step 5 of the stress computation is:
The hardening parameter is calculated using the material stress-strain curve:
σ yield (t ) = a + bε p (t )
n
EQ. 5.3.10.28
σ vm − σ yield
εɺ p ∆t =
E
where εɺ p is the plastic strain rate.
The plastic strain, or hardening parameter, is found by explicit time integration:
ε p (t + ∆t ) = ε p (t ) + εɺ p ∆t EQ. 5.3.10.29
Finally, the plastic stress is found by the method of radial return. In case of plane stress this method is
approximated because it cannot verify simultaneously the plane stress condition and the flow rule. The following
return gives a plane stress state:
σ yield el
σ ijpa = σ ij EQ. 5.3.10.30
σ vm
Iterative algorithm
Iplas=1
If flag 1 is used on card 1 of the shell property type definition, an incremental method is used. Step 5 is
performed using the incremental method described by Mendelson [1]. It has been extended to plane stress
situations. This method is more computationally expensive, but provides high accuracy on stress distribution,
especially when one is interested in residual stress or elastic return. This method is also recommended when
variable thickness is being used. After some calculations, the plastic stresses are defined as:
01-Jan-2017 45
σ xel + σ yel σ xel − σ yel

σ xpa = + EQ. 5.3.10.31
 ∆r   3∆r 
21 +  21 + 
 1− v   1+ v 
σ xel + σ yel σ xel − σ yel
σ ypa = − EQ. 5.3.10.32
 ∆r   3∆r 
21 +  21 + 
 1− v   1+ v 
σ xyel
σ xypa = EQ. 5.3.10.33
3∆r
1+
1+ v
E∆ε p
where ∆r = EQ. 5.3.10.34
2σ yield (t + ∆t )
The value of ∆ε P must be computed to determine the state of plastic stress. This is done by an iterative
method. To calculate the value of ∆ε , the von Mises yield criterion for the case of plane stress is introduced:
P
σ x2 + σ y2 − σ xσ y + 3σ xy2 = σ yield
2
(t + ∆t ) EQ. 5.3.10.35
and the values of σx, σy, σxy and σyield are replaced by their expression as a function of ∆ε P (EQS 5.3.8.31 to
5.3.8.34), with for example:
σ yield (t + ∆t ) = a + bε p (t + ∆t )
n
EQ. 5.3.10.36
and:
ε p (t + ∆t ) = ε p (t ) + ∆ε p EQ. 5.3.10.37
The nonlinear equation 5.3.10.35 is solved iteratively for ∆ε P by Newton's method using three iterations. This
is sufficient to obtain ∆ε P accurately.
5.3.10.4 Plastic plane stress with Hill’s criterion

In the case of Hill’s orthotropic criterion, the equivalent stress is given by:
σ eq2 = A1σ xx2 + A2σ yy2 − A3σ xxσ yy + A12σ xy2 EQ. 5.3.10.38
 A3 
 A1 − 2 0 
with [ A] =  A2 0 
 
 sym A12 
 
EQ. 5.3.10.14 is then written as:
[B ]{σ }n +1 = {σ ∗ } EQ. 5.3.10.39
01-Jan-2017 46
  A3   A  
1 +  A1 − 2 υ dR  A2υ − 3  dR 0 
    2

[B ] =   A1υ − A3 dR 1 +  A2 − A3 υ dR  ; dR = dε
p
E
0
 2   2   σ y 1 −υ 2
 1−υ 
 0 0 1+ A12 dR 
 2 
Changing the stress variables to {σ }:
{σ } = [Q]{σ } EQ. 5.3.10.40
with:
 A1 − A2 + C 
 1 0
 A 
 − 2 A2υ − 3  
  2  
[Q] =  A1 − A2 + C 1

0 ; C = (1 − υ )( A
2
− A2 ) + [A3 − ( A1 + A2 )υ ]
2 2
 A  1
 2 A1υ − 3  
  2  
 0 0 1
 
 
The matrix [B ] = [Q][B][Q] −1
is diagonal:
  A υ C  
1 + dR A2 − 3 − 0 0 
  2 J Q  
  
A υ C 
[B ] =  0 1 + dR A1 − 3 +
 2 J 
0  EQ. 5.3.10.41

 Q 
 
  1 − υ 
0 0 1 + dR. A12  
  2 
 
where J Q = 1 +
( A1 − A2 + C )2 is the Jacobian of [Q]. EQ. 5.3.10.40 is now written as:
 A  A 
4 A1υ − 3  A2υ − 3 
 2  2 
[B ]{σ } n −1 = {σ ∗ } EQ. 5.3.10.42
This will enable to give explicitly the expression of the yield surface EQ. 5.3.10.15:
( ) ( ) { }
f n +1 = σ eqn +1 − σ yn +1 = σ ∗ [B ] A [B ] σ ∗ − σ yn +1
2 2 t −t
[ ] −1
{ } ( ) 2
=
A11 2 A22 2 A12 2
B12
B2 B3
A
σ xx + 2 σ yy + 2 σ xy + 12 σ xxσ yy − σ yn +1
B1B2
( ) 2
EQ. 5.3.10.43
With [A ] 2×2 = [Q ]2×2 [A]2×2 [Q ]2×2 .

−t −1
01-Jan-2017 47
The derivative of f n +1 is carried out in order to use the Newton-Raphson method:
f n +1
∆ε np+1 = ∆ε np − EQ. 5.3.10.44
f n′+1
5.3.11 Calculation of forces and moments

5.3.11.1 Integration points throughout the thickness
The integration is performed using n equally spaced integration points throughout the thickness. The method
used assumes a linear variation of stresses between integrations points:
n
N x = t ∑ wkN σ xkpa EQ. 5.3.11.1
k =1
n
M x = t 2 ∑ wkM σ xkpa EQ. 5.3.11.2
k =1
Table 5.3.1 compares the coefficients used to the classical Newton quadrature in case of 3 integration points.
wN for 3 integration points wM for 3 integration points

Coefficients w1 w2 w3 Coefficients w1 w2 w3
RADIOSS 0.250 0.500 0.250 RADIOSS -0.083 0. 0.083
Simpson 0.166 0.666 0.166 Simpson -0.083 0. 0.083
Gauss Integration Scheme

Lobatto Integration Scheme
Number of
Number of Position Weight
Position Weight Points
Points
1 ±0.0000000000 2.0000000000
2 ±1.0000000000 1.0000000000
±1.0000000000 0.3333333333 2 ±1 3 1.0000000000
3
0.0000000000 1.3333333333
0.0000000000
±1.0000000000 0.1666666666 3
8/9
4
±0.4472135955 0.8333333333 ± 3 5/9
5
±1.0000000000 0.1000000000
±0.6546536707 3− 2 6 5 1
+
1
5 0.5444444444 ± 2 6 65
0.0000000000 0.7111111111 7
4
3+ 2 6 5 1 1
± −
7 2 6 65
5.3.11.2 Global plasticity algorithm

In the case of global plasticity, the forces and moments are computed directly. The algorithm is activated by
specifying the number of integration points through the thickness as zero. The first step is an obvious elastic
calculation:
{∑ } = {∑(t )}+ L{Eɺ }∆t

el
EQ. 5.3.11.3
The yield criterion used is the uncoupled Iliouchine form [13]:
N 2 16 M 2
F= + 4 − σ yield
2
≤0 EQ. 5.3.11.4
t2 t
with
01-Jan-2017 48
N 2 = N x2 + N y2 − N x N y + 3 N xy2 EQ. 5.3.11.5
M 2 = M x2 + M y2 − M x M y + 3M xy2 EQ. 5.3.11.6
where
t/2 t/2
Nx = ∫ σ x dz Mx = ∫σ
pa pa
x zdz
−t / 2 −t / 2
t/2 t/2
Ny = ∫σ My = ∫σ
pa pa
y dz y zdz
−t / 2 −t / 2
t/2 t/2
N xy = ∫ σ xy dz M xy = ∫σ
pa pa
xy zdz
−t / 2 −t / 2
An extension of Iliouchine criterion for isotropic hardening is developed here. The yield surface can be
expressed as:
 {N }  {N }
t
f =  [F ]  − σ yβ
0
( )
2
=0
{M } {M }
EQ. 5.3.11.7
1
 h 2 [A]
1
[A]
 h 
2
 2 3  sγh  
  6 
with: [F ] =   EQ. 5.3.11.8
 sym
1
2
[ A] 
  h2  
  γ  
  6 
{N }t [A]{M }
and s =
{N }t [A]{M }
EQ. 5.3.11.9
Where β and γ are scalar material characteristic constants, function of plastic deformation. They can be
identified by the material hardening law in pure traction and pure bending:
σ y (ε p )
In pure traction: f =
N2
h 2
− σ y0 β( )2
=0→β =
σ y0
EQ. 5.3.11.10
f =
M2
(
− σ y0 β )
2
=0→γ =
M
(γh / 6)
In pure bending: EQ. 5.3.11.11
2 2
σ y h2 / 6
3
σy /E+ ε p
If no hardening law in pure bending is used, γ is simply computed by γ = 2 varying between 1.0
σy / E +ε p
and 1.5.
The plasticity flow is written using the normality law:
01-Jan-2017 49
 {dε p } ∂f  {N }
 = dλ = 2dλ [F ]
{dΧ p }
 
 {N }
EQ. 5.3.11.12
∂ {M }

{M }
The equivalent plastic deformation εP is proportional to the plastic work. Its expression is the same as in the
case of traction:
{dε p }  {N }
t
σ β dε = 
0
{
p
}  = 2dλ σ y β
0
( )2
 dχ p  {M }
y EQ.5.3.11.13
This leads to:
H
dε p = 2dλσ y0 β and dβ = dε p EQ. 5.3.11.14
σ 0
y
where H is the plastic module. The derivative of function f in EQ. 5.3.11.7 is discontinuous when
{N }t {A}{M }=0. This can be treated when small steps are used by putting s=0 as explained in [87].
∂f
Then the derivative of f with respect to dε
P
( ) is carried out. The derived equation is nonlinear in
∂dε P
internal efforts and is resolved by Newton-Raphson:
 {N }  {N }  {N }
  =  − 2dλ [D ][F ]  EQ. 5.3.11.15
{M }n +1 {M }∗ {M } n +1
where [D] is the elastic stiffness matrix and:
 {N }  {N } {dε }
  =  − [D ] 
{M }* {M }n {dχ }
EQ. 5.3.11.16
5.3.12 QPH, QPPS, QEPH and QBAT shell formulations

QPH shell is the Belytschko Leviathan [17] shell for linear models or quasi-static analysis is identical to a QPPS
shell analysis, only one difference being explained in section 5.3.12.2.
The QPPS shell is a new One-point Quadrature, General Nonlinear Quadrilateral Shell Element with Physical
Stabilization. This shell is a Belytschko Leviathan [17] shell modified by Zeng and Combescure [15].
The physical stabilization is applied which enables the explicit evaluation of the stabilizing forces based on the
general degenerated shell formulation and which does not require any input parameters. An optimized choice of
the moduli is made in order to compute the stabilized forces for nonlinear material so that element's behavior is
improved with respect to similar physical stabilization elements. The cost efficiency of the element is
demonstrated by numerical examples, as compared with a fully-integrated 4-node element.
The QEPH shell is a new improved element with respect to QPH,QPPS. The improvements will be explained in
section 5.3.12.2. As the QEPH is very efficient, it replaces QPH,QPPS in the applications.
The QBAT shell is a new fully-integrated 4-node element based on Q4 γ 24 shell of Batoz and Dhatt [40] as
discussed in section 5.3.9.4.
The general formulation of the degenerated continuum quadrilateral shell (for which all these elements used) is
given in the section 5.3.12.1. The difficulties in evaluating the stabilized stiffness are also described. Section
5.3.12.2 presents the detailed formulations for the one-point quadrature shell element based on the general
formulation, and compares it with that of Belytschko and Leviathan.
01-Jan-2017 50
5.3.12.1 Formulations for a general degenerated 4-node shell

The following formulations of degenerated quadrilateral shells are based on the successful full integration
element MITC4 developed by Dvorkin-Bathe [32] and Q4 γ 24 developed by Batoz and Dhatt [40]; they are
suitable for both thin and thick shells and are applicable to linear and nonlinear problems. Their main feature is
that a classical displacement method is used to interpolate the in-plane strains (membrane, bending), and a
mix/collocation (or assumed strain) method is used to interpolate the out-plane strains (transverse shear). Certain
conditions are also specified:
• They are based on the Reissner-Mindlin model,
• In-plane strains are linear, out-plane strains(transverse shear) are constant throughout the thickness,
• Thickness is constant in the element (the normal and the fiber directions are coincident),
• 5 DOF in the local system (i.e. the nodal normal vectors are not constant from one element to
another).
A - Notational conventions
• A bold letter denotes a vector or a tensor.
• An upper case index denotes a node number; a lower case index denotes a component of vector or
tensor.
• The Einstein convention applies only for the repeated index where one is subscript and another is
superscript, e.g.:
NI xI = ∑ NI xI .
• {} denotes a vector and [ ] denotes a matrix.
B - Geometry and kinematics

Figure 5.3.11 - Coordinate systems
3
4
η
ζ Z
ξ Y
n X
1 2
a
Mid-plan
The geometry of the 4-node degenerated shell element, as shown in the Fig 5.3.11., is defined by its mid-surface
I
with coordinates denoted by Xp interpolated by the node coordinates X (I=1,4):
Xp(ξ ,η ) = N I X I EQ. 5.3.12.1
where N I (ξ ,η ) are the bilinear isoparametric shape functions, given by:

N I (ξ ,η ) = (1 + ξ I ξ )(1 + η Iη ) / 4 EQ. 5.3.12.2
A generic point q within the shell is derived from point p on the mid-surface and its coordinate along the normal
(fiber):
01-Jan-2017 51
Xq(ξ ,η , ζ ) = Xp + z ⋅ n with z = aζ
2
EQ. 5.3.12.3
where a is the shell thickness.

The transformation between the Cartesian system and the Natural system is given by the differential relation (in
matrix form):
{dXq} = ([F0 ] + z[{n,ξ } {n,η } {0}]{dξ } = [F ]{dξ } EQ. 5.3.12.4
with
[F0 ] = [{g1 }{g 2 } a2 {n}]; g1 (ξ ,η ) = X p,ξ = N I,ξ X I , g 2 (ξ ,η ) = X p,η = N I,η X I
F is the gradient tensor which is related to the Jacobian tensor [J ] = [F ]t .

With 5 DOF at each node I (three translational velocities (
v I → viI and two rotational velocities ω I → ωiI , )
the velocity interpolation is given by the Mindlin model:
v q (ξ ,η , ζ ) = v p + zβ = N I (ξ ,η )(v I + zβ I ) EQ. 5.3.12.5
β =ω×n
where β and ω are the rotational velocity vectors of the normal: β = β1 t 1 + β 2 t 2 = ω 2 t 1 − ω1 t 2
and (t1, t 2 , n ) is base of the local coordinate system.
EQ. 5.3.12.5 can be written also by:
( )
v i = N I v iI + z − N Iω1I t I2i + N Iω 2I t 1iI ; i = 1,3
This velocity interpolation is expressed in the global system, but ωI must be defined first in the local nodal
coordinate system to ensure Mindlin's kinematic condition.
C - Strain-rate construction
The in-plane rate-of-deformation is interpolated by the usual displacement method.
The rate-of-deformation tensor (or velocity-strains) D = L t + L t ( T

)/ 2 is defined by the velocity gradient tensor
L:
{dv } = [L ]{dξ } = [L ][F ] {dXq} = [L]{dXq} = [Q][L ][Q ] {dXq}

q ζ ζ
−1
t
T
EQ. 5.3.12.6
with [Q] = [{t1}{t2 }{n}] .

The Reissner-Mindlin conditions εζ = 0 and σζ = 0 requires that the strain and stress tensors are computed in
the local coordinate system (at each quadrature point).
After the linearization of Lt with respect to z, the in-plane rate-of-deformation terms are given by:
[Lt ]2×2 = [Ls 0 ] + z[Ls1 ]

with the membrane terms:
[Ls 0 ] = [C1 ][C0 ]
the bending terms:
01-Jan-2017 52
 g 2 ⋅ n,η − g1 ⋅ n,η   t 1 ⋅ β, ξ t1 ⋅ β,η 

[Ls1 ] = 2H[Ls0 ] + [C1 ] 2 [C0 ] + 
t1 ⋅ β,η 
EQ. 5.3.12.7
− g ⋅ n,ξ g ⋅ n,ξ   t 1 ⋅ β, ξ
1
where the contravariant vectors g α (α = 1,2) , dual to g α , satisfy the orthogonality condition: g α ⋅ g β = δ βα
(Kronecker delta symbol); H (ξ ,η ) is the average curvature: 2H = − g 1⋅ n, ξ +g 2 ⋅ n,η ( )
 t 1 ⋅ v p ,ξ t 1 ⋅ v p ,η 
[C0 ] =  t1 ⋅ g2 t 2 ⋅ g1 
 1
2
and [C1 ] = 
t 2 ⋅ v p ,η 
EQ. 5.3.12.8
 t1 ⋅ g t2 ⋅ g   t 2 ⋅ v p ,ξ
The curvature-translation coupling is presented in the bending terms for a warped element (the first two terms in
last EQ.)
The out-plane rate-of-deformation (transverse shear) is interpolated by the “assumed strain” method, which is
based on the Hu-Washizu variation principle.
If the out-plane rate-of-deformation is interpolated in the same manner for a full integration scheme, it will lead
to “shear locking”. It is known that the transverse shear strains energy cannot vanish when it is subjected to a
constant bending moment. Dvorkin-Bathe's [32] mix/collocation method has been proved very efficient in
overcoming this problem. This method consists in interpolating the transverse shear from the values of the
covariant components of the transverse shear strains at 4 mid-side points. i.e.:
[
γ ξ = (1 − η )γ ξ A1 + (1 + η )γ ξ A2 / 2 ] EQ. 5.3.12.9
γη = [(1 − ξ )γ + (1 + ξ )γ η ]/ 2
B1 B2
η EQ. 5.3.12.10
Where γ ξ Aα , γ η Bα are the values of the covariant components at 4 mid-side points which vanish under a
constant bending moment (see Figure 5.3.12).
Figure 5.3.12 - Covariant components at 4 mid-side
A2
γξ
4
A2
B1 γη 3
γη
γξ
B1 B2 B2
γη
A1
1 2
A1
γξ
D - Special case for one-point quadrature and the difficulties in

stabilization
The formulations described above are general for both the full integration and reduced integration schemes. For a
one-point quadrature element, you have the following particularities:
The quadrature point is often chosen at (ξ = 0,η = 0) . The derivatives of the shape functions are:
N I ,ξ = (ξ I + hIη ) / 4 N I ,η = (η I + hI ξ ) / 4 EQ. 5.3.12.11
Where hI = (1 − 1 1 − 1) .
01-Jan-2017 53
This implies that all the terms computed at the quadrature point are the constant parts with respect to (ξ ,η ) , and
the stabilizing terms (hourglass) are the non-constant parts.
The constant parts can be derived directly from the general formulations at the quadrature point without
difficulty. The difficulties in stabilization lie in correctly computing the internal force vector (or stiffness
matrices):
f int = ∫ B t ⋅ σ dΩ = f int (ξ = η = 0 ) + f stab

int
EQ. 5.3.12.12
Ω
int
It would be ideal if the integration term f stab could be evaluated explicitly. But such is not the case, and the
main obstacles are the following:
For a non-coplanar element, the normal varies at each point so that it is difficult to write the non-constant part of
strains explicitly. For a physically nonlinear problem, the non-constant part of stress is not generally in an
explicit form. Thus, simplification becomes necessary.
5.3.12.2 Fully-integrated shell element QBAT

QBAT is a fully-integrated shell element based on the general formulation described above. In the surface of
each element, 4 Gauss points are used to evaluate the nodal forces.
The main modifications with respect to Q4 γ 24 shell element [40] are the following:
• Reduced integration for in-plane shear (constant) to avoid locking.
• Co-rotational coordinate system is used and the stresses are evaluated in 4 local systems at each Gauss
points..
5.3.12.3 The new one-point quadrature shell element

In this section, a one-point quadrature shell element formulation will be developed from the general formulation
described in the previous section. It is based on the Physical Stabilization method which explicitly computes the
stabilization terms in making some simplifications.
The following formulations will be written in the local coordinate system [t1 t2 n] (the circumflex in the co-
rotational system notation has been omitted for convenience).
A - Kinematic approximation
I I
The velocity interpolation using the nodal tangent vectors ( t1 , t 2 ) complicates the strain computation, especially
for transverse shear which is used mainly as a penalty function. To be consistent with the one-point quadrature
approach, the kinematic approximation is performed by:
[ ]
vi = N I viI + z (− δ i 2ω1I + δ i1ω 2I ) EQ. 5.3.12.13
where ωi I (i=1,2) is the nodal rotation velocity around t iI . ω i I can be computed by a projection scheme by :
ω I = ω I − (ω I ⋅ n I )n I = P(n I )ω I EQ. 5.3.12.14

The projection consists in eliminating the nodal drilling rotations in order to reinforce Mindlin's kinematic
condition at the nodes. It has been pointed out [18] that without this projection, the element is too flexible and
cannot pass the Twisted Beam test.
This projection has a drawback of changing the invariance property to rigid body motion, i.e.: a warped element
being invariant to rigid body rotation will now strains under to rigid body rotation if the drill projection is
applied. To overcome this problem, a full projection proposed by Belytschko-Leviathan [66] which free either
drilling rotation or rigid motion should be used. This full projection is only used for QEPH element.
01-Jan-2017 54
B - In-plane strain-rate construction

Constant part
It is useful to write the shape functions in Belytschko-Bachrach's mixed form:
N I ( x, y, ξη ) = ∆ I + bxIx + b yIy + γ I φ EQ. 5.3.12.15
with: [ ]
∆ I = t I − (t I x I ) bxI − (t I y I )b yI ; t I = (1,1,1,1)
b xI = ( y 24 y 31 y 42 y13 ) / A; b yI = ( x 42 x13 x 24 x31 ) / A ( fij = ( fi − fj ) / 2)

γ I = [hI − (hJ x J )bxI − (hJ y J )b yI ]/ 4 ; φ = ξη ;
A is the area of element.

The derivation of the shape functions is given by:
N I ,α = bαI + γ I φ,α (α = x, y )
where:
φ , x = ηξ , x + ξη , x = J −1η + J 12−1ξ
11
EQ. 5.3.12.16
φ , y = ηξ , y +ξη , y = J 21−1η + J 22−1ξ
The advantage of this shape function form is that a linear field expressed with Cartesian coordinates and a
bilinear field expressed with Natural coordinates is decomposed so that the constant part is directly formulated
with the Cartesian coordinates, and the non-constant part is to be approached separately.
The in-plane rate-of-deformation (decomposed on membrane and bending) is given by:
{D} = {D m }+ z{Db } = [BI ]{vnI } EQ. 5.3.12.17
with:
BI  = [BI ] + z[BI ] ; < vn >=< vx v y vzϖ xϖ y > .

m b I I I I I I
The development of the general formulations leads to the constant part, denoted by superscript 0, of the matrix
[BI]:
bxI 0 0 bxIc 0 0 0 bxI 

[(B ) ]
m 0
I

= 0 byI 0

[(B ) ]= 2H [(B ) ]
b 0 m 0
I

+ 0 b c
yI 0 − byI

0  EQ. 5.3.12.18
byI bxI 0 byIc b c
0 − bxI byI 
  xI
with
< bxIc >= 4 z γ / A 2 < b y 2 ; b y1 ; b y 4 ; b y 3 >
< b yIc >= 4 z γ / A 2 < bx 4 ; bx 3 ; bx 2 ; bx1 >
The parameter zγ = γ I z I is a measure of the warping of the element.
01-Jan-2017 55
The first term [ ]

2 H (BIm ) is neglectable. You have verified that the order of 2 H (BIm ) is ε times the
0
[ ] 0
second term of [(B ) ]

b 0
with ε = 2 x34 / y34 (f = (f
ij i + f j )/ 2) , which vanishes when the element is
rectangular. Thus, this term is not used in the program.
The constant part of the in-plane rate-of-deformation formulation without the H term is consistent with the result
of Belytschko's family shell element [24], [17], though this part has been obtained in a very different manner.
Letellier has given the same result in his thesis [43], and studies were also made of the quadratic terms with
respect to z.
Non-constant part
The main simplification for the non-constant part formulation, in order to overcome the difficulties described
above, is the following:
The element is considered to be flat.
In this case, the Jacobian matrix is written as:
 x, ξ 
 = [J ]
y, ξ 0
0 
[J ] = [F0 ] =  x,η
t
y,η 0   0 aJ / 2 J  EQ. 5.3.12.19
 0 0 aJ 0 / 2 J   0 
with the determinant J of the in-plane Jacobian:

J = det[J ] = J 0 + J1ξ + J 2η EQ. 5.3.12.20
and:
[
J 0 = (ξ I x I )(η I y I ) − (η I x I )(ξ I y I ) / 16 = A / 4 ]
[ ]
J1 = (ξ I x I )(hI y I ) − (hI x I )(ξ I y I ) / 16
J2 = [(h x )(η y ) − (η x )(h y )]/ 16

I
I
I
I
I
I
I
I
The inverse of the in-plane Jacobian matrix can be expressed explicitly:
ξ , x η , x  1  y,η − y,ξ  1 (η I + hI ξ ) y I (ξ I + hIη )y I 

[J ] =  =  =
x, ξ  4 J  (η I + hI ξ )x I (ξ I + hIη )x I 
EQ. 5.3.12.21
ξ , y η , y  J − x,η
You can now write the non-constant part, denoted by superscript H, of the matrix [BI] for in-plane rate-of-
deformation:
γ I φ , x 0 0  0 γ Iφ,x 
[(B ) ]
m H
I

= 0 γ Iφ, y

0 ; [(B ) ]
b H
I

= − γ I φ , y 0 

EQ. 5.3.12.22
γ I φ , y γ Iφ, x 0 − γ I φ , x γ I φ , y 
 
[(B ) ]
It is shown in [16] that the non-constant part of membrane rate-strain does not vanish when a warped element
m H
undergoes a rigid body rotation. Thus, a modified matrix I is chosen:
γ I φ , x 0 z γ b xI φ , x 
[(B ) ] m H
I

= 0 γ Iφ, y z γ b yI φ , y

 EQ. 5.3.12.23
γ I φ , y
 γ Iφ, x z γ (b xI φ , y +b yI φ , x )
This matrix is different from the Belytschko-Leviathan correction term added at rotational positions, which
couples translations to curvatures as follows:
01-Jan-2017 56
 1 
γ I φ , x 0 0 0 −
4
zγ φ , x 
[(B ) ]
m H
I

= 0

γ Iφ, y 0
1
4
zγ φ , y 0



EQ. 5.3.12.24
γ φ , γ Iφ,x 0
1
zγ φ , x − zγ φ , y 
1
 I y 4 4 
This will lead to 'membrane locking' (the membrane strain will not vanish under a constant bending loading).
According to the general formulation, the coupling is presented in the bending terms not in the membrane terms,
yet the normal translation components in (BIm)i3 do not vanish for a warped element due to the tangent vectors
ti (ξ ,η ) which differ from ti (0,0).
C - Out-plane strain-rate construction

The out-plane rate-of-deformation (transverse shear) is interpolated by the Dvorkin-Bathe method, whose closed
form is given by Belytschko-Leviathan:
 v zI 
 Dxz   I
  = [BIc ]ϖ x  EQ. 5.3.12.25
D
 yz  ϖ I 
 y
where: [BIc ](ζ ,η ) = [BIcη ](η ) + [BIcξ ](ξ )
1  4η I y ξ I − (ξ J + hJη I )η k y k y J (ξ J + hJη I )ηk y k x J 
(B )
I
η 0
=  
Ic
16 A − 4η I x I ξ I (ξ J + hJη I )ηk x k y J − (ξ J + hJη I )η k x k x J 
1 − 4ξ I y η I (η J + hJ ξ I )ξ k y k y J − (η J + hJ ξ I )ξ k y k x J 
(B )
I
ξ 0
= 
(η J + hJ ξ I )ξk x k x J 
EQ. 5.3.12.26
16 A  4ξ I y Iη I − (η J + hJ ξ I )ξ k x k y J
Ic
and (B )Ic
η H
= η I BIc( ) η ; (B )
η 0
Ic
ξ H
( )ξ
= ξ I BIc
ξ 0
The straightforward form of [BIc] is obtained using one additional simplification:
ξ, x (ξ ,η ) = ξ, x (0,0 )η, x (ξ ,η ) = η, x (0,0)

which is true for a parallelogram element. Although this simplification is not necessary, it is justified by the fact
that the transverse shear terms serve mainly as a penalty function.
D - Explicit integration of the nodal internal force vector

Elastic case
The elementary nodal force vector is computed by:
{f } = ∫ [B ] [C ][B ]{v }dv

I
int
ve
I
t
J
J
n
e
Taking advantage of substantial orthogonality between:

• the constant in-plane fields along with the non-constant ones
• the membrane and the bending
• the in-plane fields and the out-plane fields
• the decomposed non-constant out-plane fields
01-Jan-2017 57
resulting in:
{f } = {( f ) }+ {( f ) }
I
int
I
int 0
I
int H
With {( f ) } the constant part being computed with one-point quadrature, and
int 0
I
{( f ) }= ∫ [(B ) ] [C ][(B ) ]{v }dv

I
int H
ve
I
H t
J
H J
n
e
It can be shown in the last equation that only the following scalar functions need to be integrated:
H xx = ∫ Aeφ, xφ, x dAe , H xy = ∫ Aeφ, xφ, y dAe , H yy = ∫ Aeφ, yφ, y dAe EQ. 5.3.12.27
These can be evaluated explicitly.

Defining 6 hourglass generalized rate-of-deformation qɺ by:
 qɺ xm = γ I v xI + z1bxI vzI
membrane :  m EQ. 5.3.12.28
 qɺ y = γ I v y + z byI v z
I 1 I
 qɺ xb = γ Iϖ yI
bending :  b EQ. 5.3.12.29
 qɺ y = γ Iϖ x
I
 qɺ xs = 4hI v zI − (ξ I y I )η I + ( hI y I ) t I ϖ xI + (ξ I x I )ηI + ( hI x I ) t I ϖ yI

    
shear :  EQ. 5.3.12.30
 qɺ y = 4hI vz −  (η I y ) ξ I + ( hI y ) t I ϖ x +  (η I x ) ξ I + ( hI x ) t I ϖ y
s I I I I I I I
The rate-of-deformations will be written explicitly.

The rate form of the constitutive relation is written as (stress plane for in-plane terms):
σɺ = C:D
With the assumption: the spin is constant within the element, the objectivity principle will be satisfied. The
incremental computation is performed with the hourglass generalized rate-of-deformation qɺ , i.e.:
(q )n +1 = (q )n + qɺn + 1 ∆t
2
Noting that if φ ,α is considered as constant over a time step, it is equivalent to the incremental stress
computation.
Physical nonlinear case

You will now consider an elastoplastic problem.
The elementary nodal internal force vector is now computed by:
{f } = ∫ [B ] {σ }dv
I
int
ve
I
t e
The constitutive relation is written by either a tangent form: σɺ ( )

= C t:D , or a projection form: σ = P σ e ,... ,
t
( ) { }
e
[ ]
where C σ ,... is the history-dependent tangent tensor ; σ n +1 = {σ n } + C {Dn +1 / 2 }∆t is the trial stress,
and the function P consists of projecting the trial stress on the updated yield surface.
The decomposed form of the last equation is written:
{f } = {( f ) }+ {( f )
I
int
I
int 0
I
int H
}
01-Jan-2017 58
The constant part is computed by integration at each integration point through the thickness.
The stabilization part will be approached by relying on two hypotheses:
• Keep the same orthogonalities as in the elastic case, and
Use the unidimensional tangent modulus Et (ζ ) to evaluate the non-constant rate-stress, i.e.:
{σɺ }H = (Et (ζ ) / E )[C ]{D}H EQ. 5.3.12.31
where E is the Young's modulus and [C ] is the matrix of elastic moduli.

Thus, the elastic case easily extends to the nonlinear case.
The incremental computation with the hourglass generalized rate-of-deformation qɺ becomes:
( ) = (q ) + λ qɺ
membrane : q m n +1
m
n m
m
n +1 / 2 ∆t EQ. 5.3.12.32
bending : (q ) = (q ) + λ qɺ
b
n +1 ∆t b
n b
b
n +1 / 2 EQ. 5.3.12.33
shear : (q ) = (q ) + qɺ
s
n +1 ∆t s
n
s
n +1 / 2 EQ. 5.3.12.34
Where Et (ζ ) is obtained by the constant stress incremental computation along the thickness and Et (ζ ) = E
in the elastic zone, and λm = Et / E; Et is the average value of Et (ζ ) and λb = min Et (ζ ) / E .
For the QPH shell, λm = 1

The key orthogonalities has been maintained without any significant deterioration in performance, although the
first two orthogonalities might have been slightly violated. In fact, it is simply due to these orthogonalities that a
one-point quadrature element dramatically reduces the computation cost; otherwise you return to the full
integration scheme.
Most of the physical stabilization elements have incorporated the following assumption:
The material response is constant within the element.
There are two alternatives to this assumption:
• to take the elastic matrix [C ] directly:

{σɺ } = {σɺ }0 + [C ]{(εɺ − εɺ P )H }= {σɺ }0 + [C ]{(εɺ )H }→ {(εɺ P )H }= 0
which means that the plastic rate-of-deformation {εɺ } is constant within the element; or
P
• to take the tangent matrix [Ct] ( ξ = η = 0 , ζ = constant):

Since the components of [Ct] are generally functions of the updated stress (precisely, the stress deviator for an
elastoplastic problem with an associative flow rule, which means that the stress is constant within the element.
Neither of these alternatives is theoretically perfect. Let us note that the [C] option results in a contradiction with
the stress computation (which yields different results for the constant part and the non-constant part); it is more
expensive and the tangent form is not generally used for constant stress computations within an explicit program.
Hence, the approximation based on the above assumption is not necessary.
The choice of the moduli for the nonlinear case has not been studied for Belytschko-Leviathan's element [17],
and it has been shown that this choice has little effect on the result of the “Cylindrical panel test”. In [30], the
elastic tangent matrix has been used for the evaluation of the stabilizing forces.
QPH,QPPS have shown often stiffer behaviors and sometimes have certain numerical problems in crash
simulations.
01-Jan-2017 59
5.3.12.4 Advanced elasto-plastic hourglass control

QEPH (Quadrilateral ElastoPlastic Physical Hourglass Control) element
With one-point integration formulation, if the non-constant part follows exactly the state of constant part for the
case of elasto-plastic calculation, the plasticity will be under-estimated due to the fact that the constant
equivalent stress is often the smallest one in the element and element will be stiffer. Therefore, defining a yield
criterion for the non-constant part seems to be a good idea to overcome this drawback.
From EQ. 5.3.12.16 and EQ. 5.3.12.21, you have the rate of stresses of non-constant part:
φ, x qɺ αx  σɺ xαη  σɺ xαξ 

     
{σɺ i }H = [C ]{εɺ}H = [C ]φ , y qɺ αy  = η σɺ αyη  + ξ σɺ αyξ  EQ. 5.3.12.35
 0   0   0 
     
Where α = m, b corresponds to the membrane and bending terms respectively. Note that the shear terms are
eliminated to avoid shear locking. The transverse shear terms can also be written as the same way:
τɺ xη  τɺ xξ 
{τɺi }H = η  +ξ  EQ. 5.3.12.36
τɺ yη  τɺ yξ 
You can now redefine 12 generalized hourglass stresses by integrating their rate ones, and the stress field can be
expressed by:
σ αxη  σ xαξ 
Membrane, bending: {σ } = {σ } + {σ } = {σ }
α
i
α 0
i
α H
i
α 0
i
 α   
+ η σ yη  + ξ σ αyξ 
 0   0 
   
τɺ xη  τɺ xξ 
Shear: {τ i } = {τ i }0 + {τ i }H = {τ i }0 + η  +ξ 
τɺ yη  τɺ yξ 
Even the redefinition for shear is not necessary as it is not included in the plastic yield criterion, but the same
stress calculation as the constant part with the updated Lagrangian formulation is always useful when large strain
is involved.
Plastic yield criterion:

The von Mises type of criterion is written by:
f = σ eq2 (ξ ,η , ζ ) − σ y2 = 0 EQ. 5.3.12.37
for any point in the solid element, where σy is evaluated at the quadrature point.
As only one criterion is used for the non-constant part, two choices are possible:
f = f (σ eq ); σ eq =
1
σ eq dΩ
Ω Ω∫
• taking the mean value, i.e.:
• taking the value by some representative points, e.g. eight Gausse points
The second choice has been used in this element.
01-Jan-2017 60
Elasto-plastic hourglass stress calculation:

The incremental hourglass stress is computed by:
• Elastic increment
(σ i )trH = (σ i )n + [C ]{εɺ} ∆t
H H
n +1
• Check the yield criterion

• If f ≥ 0 , the hourglass stress correction will be done by un radial return
(σ i )nH+1 = P((σ i )trH

n +1
,f)
5.3.13 Three-node shell elements

As for the four node shell element, a simple linear Mindlin Plate element formulation is used. Likewise, the use
of one integration point and rigid body motion given by the time evolution of the local reference frame is
applied. There is no hourglass mode in case of one integration point.
5.3.13.1 Local Reference Frame

The local reference frame for the three node shell element is shown in Figure 5.3.13.
Figure 5.3.13 - Node Shell Local Reference Frame
The vector normal to the plane of the element is defined as:
g1 × g 2
n= EQ. 5.3.13.1
g1 × g 2
The vector defining the local x direction is defined as edge 1-2:
g1
t1 = EQ. 5.3.13.2
g1
01-Jan-2017 61
Hence, the vector defining the local y direction is found from the cross product of the two previous vectors:
t 2 = n × t1 EQ. 5.3.13.3
5.3.13.2 Time Step

The characteristic length for computing the critical time step is defined by:
2area
L=
(
max 12, 23, 31 ) EQ. 5.3.13.4
Three Node Shell Shape Functions

The three node shell has a linear shape functions defined as:
φ1 = a1 + b1 x + c1 y EQ. 5.3.13.5
φ2 = a2 + b2 x + c2 y EQ. 5.3.13.6
φ3 = a3 + b3 x + c3 y EQ. 5.3.13.7
These shape functions are used to determine the velocity field in the element:
3
vx = ∑ φI v xI EQ. 5.3.13.8
I =1
3
v y = ∑ φ I v yI EQ. 5.3.13.9
I =1
3
v z = ∑ φ I v zI EQ. 5.3.13.10
I =1
3
ω x = ∑ φ I ω xI EQ. 5.3.13.11
I =1
3
ω y = ∑ φI ω yI EQ. 5.3.13.12
I =1
∂v x 3
∂φ
= ∑ I v xI EQ. 5.3.13.13
∂x I =1 ∂x
∂v x 3
∂φ
= ∑ I v xI EQ. 5.3.13.14
∂y I =1 ∂y
5.3.13.4 Membrane Behavior

The method used to calculate the membrane behavior and the membrane strain rates is exactly the same as that
used for four node shell elements (see section 5.3.6.1).
5.3.13.5 Bending Behavior

The bending behavior and calculation of the bending strain rates (or curvature rates) is the exact same method
used for four node shell elements (see section 5.3.6.2).
01-Jan-2017 62
5.3.13.6 Strain Rate Calculation

The strain rate calculation for the three node shell is the same as the method used for the four node shell.
However, only three nodes are accounted for. This makes the vectors and matrices smaller. The overall
membrane strain rate is calculated by:
{eɺ}m = {eɺx , eɺ y ,2eɺxy } EQ. 5.3.13.15
{v}m = {v1x , v1y , vx2 , v 2y , vx3 , v3y } EQ. 5.3.13.16
{eɺ}m = [B ]m {v}m EQ. 5.3.13.17
where the [B]m matrix of shape function gradients is defined as:

 ∂φ1 ∂φ2 
 0 0 0 0 
 ∂x ∂x 
∂φ1 ∂φ2 ∂φ3 
[B ]m = 0

0 0 EQ. 5.3.13.18
∂y ∂y ∂y 
 ∂φ ∂φ1 ∂φ2 ∂φ2 ∂φ3 ∂φ3 
 1 
 ∂y ∂x ∂y ∂x ∂y ∂x 
∂φ3
where = 0 for a shell element.
∂x
The overall bending strain or curvature rate is computed by:
{eɺ}b = {kɺx , kɺy ,2kɺxy ,2eɺzx ,2eɺyz } EQ. 5.3.13.19
{v}b = {ω1y ,−ω1x , ω y2 ,−ωxɺ2ω y3 ,−ωx3 , v1z , vz2 , vz3 } EQ. 5.3.13.20
{eɺ}b = [B ]b {v}bm EQ. 5.3.13.21
where:
 ∂φ1 ∂φ2 
 ∂x 0 0 0 0 0 0 0 
∂x
 ∂φ1 ∂φ2 ∂φ3 
 0 0 0 0 0 0 
 ∂y ∂y ∂y 
[B ] =  ∂φ1 ∂φ1 ∂φ2 ∂φ2 ∂φ3 ∂φ3
 
0 0 0  EQ. 5.3.13.22
 ∂y ∂x ∂y ∂x ∂y ∂x 
φ ∂φ1 ∂φ2 ∂φ3 
0 φ2 0 φ3 0
 1 ∂x ∂x ∂x 
 ∂φ1 ∂φ2 ∂φ3 
 0 φ1 0 φ2 0 φ3 
 ∂y ∂y ∂y 
5.3.13.7 Mass and Inertia

The three node shell element is considered as an element with a lumped mass. Its mass is defined as:
m = ρAt EQ. 5.3.13.23
where:
ρ is the material density,
t is the shell thickness,
01-Jan-2017 63
A is the reference plane surface area.

The mass moment of inertia about all axes is the same:
 2A t 2 
I xx = m +  EQ. 5.3.13.24
 6 12 
I zz = I yy = I xx EQ. 5.3.13.25
I xy = 0 EQ. 5.3.13.26
When nodal masses need to be calculated, the distribution is determined by the shape of the element as shown in
Figure 5.3.14.
Figure 5.3.14 - Mass distribution
α3
α1 α2
2
1
The mass and inertia at node i are given by:

αi α
mi = m ; Ii = i I EQ. 5.3.13.27
π π
5.3.14 Composite Shell Elements

There are three different element types that can be used for modeling composites. These are:
• Type 9 Element Property - Orthotropic Shell
• Type 10 Element Property - Composite Shell
• Type 11 Element Property - Composite Shell with variable layers
These elements are primarily used with the Tsai-Wu model (material law 25). They allow one global behavior or
varying characteristics per layer, with varying orthotropic orientations, varying thickness and/or varying material
properties, depending on which element is used. Elastic, plastic and failure modeling can be undertaken.
5.3.14.1 Transformation matrix from global to orthotropic skew

If the element reference is defined by the axes X, Y and the orthotropy directions by axes 1-2, write:
Figure 5.3.15 - Fiber orientation
01-Jan-2017 64
c − s 
 s c  with c =cosθ and s = sinθ.
 
ε 11 ε 12   c s   ε X ε XY  c − s 
ε = ⋅ ⋅
ε Y   s c 
EQ. 5.3.14.1
 12 ε 22  − s c  ε XY
ε11 ε12   c s  c ⋅ ε X + s.ε XY − s ⋅ ε X + c.ε XY 
ε = ⋅ − s ⋅ ε XY + c.ε Y 
EQ. 5.3.14.2
 12 ε 22  − s c   c ⋅ ε XY + s.ε Y
ε11 ε12   c 2 ⋅ ε X + 2cs.ε XY + s 2 .ε Y ( )
− cs ⋅ ε X + c 2 − s 2 .ε XY + cs.ε Y 
=
ε
( )  EQ. 5.3.14.3
 12 ε 22  − cs ⋅ ε X + c − s .ε XY + cs.ε Y s 2 .ε X − 2cs.ε XY + c 2 .ε Y
2 2

The strain-stress relation in orthotropy directions is written as:
{σ } = [C ]{ε } EQ. 5.3.14.4
−ν 21
 E1 E22 0
[C ]−1 =  −Eν1112 
11
1
E22 0 EQ. 5.3.14.5
 0 0 1 
 G12 
σ = σ 11 σ 22 σ 12 ; ε = ε 11 ε 22 ε 12 EQ. 5.3.14.6
The computed stresses are then projected to the element reference:
σ XX σ XY  c − s  σ 11 σ 12   c s 
σ = ⋅ ⋅
σ YY   s c  σ 12 σ 22  − s c 
EQ. 5.3.14.7
 XY
5.3.14.2 Composite modeling in RADIOSS

RADIOSS has been successfully used to predict the behavior of composite structures for crash and impact
simulations in the automotive, rail and aeronautical industries.
The purpose of this chapter is to present the various options available in RADIOSS to model composites, as well
as some modeling methods.
Modeling composites with shell elements

Composite materials with up to 100 layers may be modeled, each with different material properties, thickness,
and fiber directions.
Lamina plasticity is taken into account using the Tsai-Wu criteria, which may also consider strain rate effects.
Plastic work is used as a plasticity as well as rupture criterion.
Fiber brittle rupture may be taken into account in both orthotropic directions.
Delamination may be taken into account through a damage parameter in shear direction.
Modeling composites with solid elements

Solid elements may be used for composites.
Two material laws are available:
- Solid composite materials: one layer of composite is modeled with one solid element. Orthotropic
characteristics and yield criteria are the same as shell elements (see above).
- A honeycomb material law is also available, featuring user defined yield curves for all components of the stress
tensor and rupture strain.
01-Jan-2017 65
Solid + shell elements

For sandwich plates, if the foam or the honeycomb is very thick, it is possible to combine composite shells for
the plates and solid elements for the sandwich (see figure below).
5.3.14.3 Element orientation

A global reference vector v is used to define the fiber direction. The direction in which the material properties
(or fiber direction) lay is known as the direction 1 of the local coordinate system of orthotropy. It is defined by
the Φ angle, which is the angle between the local direction 1 (fiber direction) and the projection of the global
vector v as shown in Figure 5.3.16.
Figure 5.3.16 - Fiber Direction Orientation
The shell normal defines the positive direction for φ . For elements with more than one layer, multiple φ angles
can be defined.
The fiber direction orientation may be updated by two different ways:
1. constant orientation in local corotational reference frameconstant orientation in local isoparametric
frame. The first formulation may lead to unstable models especially in the case of very thin shells (e.g.
airbag modelling). In Figure 5.3.17 the difference between the two formulations is illustrated for the case of
element traction and shearing.
01-Jan-2017 66
Figure 5.3.17 - Fiber Direction Updating
Formulation 1
Shear
Formulation 2
Formulation 1 F
α
F
Traction α
Formulation 2
5.3.14.4 Orthotropic shells

The type 9 element property set defines orthotropic shell elements. They have the following properties:
1. Only one layer.
2. Can have up to 5 integration points1 through the thickness.
3. One orientation.
4. One material property.
5.3.14.5 Composite shell

The type 10 element property set defines composite shell elements. They have the following properties:
1. Up to 100 layers can be modeled.
2. Constant layer thickness.
3. Constant reference vector.
4. Variable layer orientation.
5. Constant material properties.
Integration is performed with constant stress distribution for each layer.
01-Jan-2017 67
5.3.14.6 Composite shell with variable layers

The type 11 element property set defines composite shell elements that allow variable layer thicknesses and
materials. They have the following properties:
1. Up to 100 layers can be modeled.
2. Variable layer thicknesses.
3. Constant reference vector.
4. Variable layer orientation.
5. Variable material properties, mi2,3
Integration is performed with constant stress distribution for each layer.
1. Same integration rule as shells.

2. Material number m i must be defined as law type 25 or type 27 (not both) in material input cards.
3. Material given in the shell definition is only used for memory allocation, time step computation and interface
stiffness. It must also be defined as law 25.
5.3.14.7 Limitations
When modeling a composite material, there are two strategies that may be applied. The first, and simplest, is to
model the material in a laminate behavior. This involves using type 9 property shell elements. The second is to
model each ply of the laminate using one integration point. This requires either a type 10 or 11 element.
Modeling using the type 9 element allows global behavior to be modeled. Input is simple, with only the
reference vector as the extra information. A Tsai-Wu yield criterion and hardening law is easily obtained from
the manufacturer or a test of the whole material.
Using the type 10 or 11 element, one model’s each ply in detail, with one integration point per ply and tensile
failure is described in detail for each ply. However, the input requirements are complex, especially for the type
11 element.
Delamination is the separation of the various layers in a composite material. It can occur in situations of large
deformation and fatigue. This phenomenon cannot be modeled in detail using shell theory. A global criterion is
available in material law 25. Delamination can affect the material by reducing the bending stiffness and buckling
force.
5.3.15 Three-node triangle without rotational d.o.f. (SH3N6)

The need of simple and efficient element in nonlinear analysis of shells undergoing large rotations is evident in
crash and sheet metal forming simulations. The constant-moment plate elements fit this need. One of the famous
concepts in this field is that of Batoz et al. [40] known under DKT elements where DKT stands for Discrete
Kirchhoff Triangle. The DKT12 element [40], [105] has a total of 12 d.o.f’s. The discrete Kirchhoff plate
conditions are imposed at three mid-point of each side. The element makes use of rotational d.o.f. at each edge to
take into account the bending effects. A simplified three-node element without rotational d.o.f. is presented in
[106]. The rotational d.o.f. is computed with the help of out-of-plane translational d.o.f. in the neighbor
elements. This attractive approach is used in RADIOSS in the development of element SH3N6 which based on
DKT12.
01-Jan-2017 68
5.3.15.1 Strain computation

Consider two adjacent coplanar elements with a common edge i-j as shown in Figure 5.3.19. Due to out-of-
plane displacements of nodes m and k, the elements rotate around the side i-j. The angles between final and
initial positions of the elements are respectively α m and α k for corresponding opposite nodes m and k.
Assuming, a constant curvature for both of elements, the rotation angles θm and θk related to the bending of
each element around the common side are obtained by:
hk hm
θk = and θm = EQ.5.3.15.1
2R 2R
However, for total rotation you have:
θk + θm = αk + αm EQ.5.3.15.2
which leads to:
θk =
(α k + α m )hk and θm =
(α k + α m )hm EQ.5.3.15.3
(hk + hm ) (hk + hm )
Figure 5.3.18 – Computation of rotational d.o.f. in SH3N6
hm m
k
hk
i
O
θk θm
αm
αk
i, j
01-Jan-2017 69
Consider the triangle element in Figure 5.3.19. The outward normal vectors at the three sides are defined and
denoted n1, n2 and n3. The normal component strain due to the bending around the element side is obtained
using plate assumption:
 α1 
α 
ε n1   (h1 +2 h5 ) 0 0 0 2
( h1 + h5 ) 0  2 
    α3 
ε n 2  =  0
2
( h2 + h6 )   
2
( h2 + h6 ) 0 0 0 EQ.5.3.15.4
γ   0 α
 n3   0 2
( h3 + h4 )
2
(h3 + h4 ) 0 0   4 
α 5 
 
α 6 
The six mid-side rotations α i are related to the out-of-plane displacements of the six apex nodes as shown in
Figure 5.3.20 by the following relation:
 α1   h1
1
− cosh2β 3 − cosh3β 2 0 0 0   w1 
α   − cos β 3 1
− cos β1
0 0

0  w 2 
 2   cosh1β h2 h3  
α 3  − h1 2 − cos β1 1
0 0 0   w 3 
  =  cosγ 2  
h2 h3
cos γ 1 EQ.5.3.15.5
α 4   − q1 − 0  w 4 
1
q2 0 h4 0
α 5   0 − cosψ 3
− cosr3ψ 2 0 1
0  w 5 
   cosφ2 r2 h5
 
α 6   − s
 1
0 − coss3φ1 0 0 1  w 
h6   6 
(
where (h1 h2 h3 ) , q1 h4 q2 , r2 ) ( h5 r3 ) and (s3 h6 s1 ) are respectively the heights of the
triangles (1,2,3), (1,4,2), (2,5,3) and (3,6,1).
Figure 5.3.19 – Normal vectors definition
n1 2
n3
3
n2
1
01-Jan-2017 70
The non-null components of strain tensor in the local element reference are related to the normal components of
strain by the following relation (see [40] and [106] for details):
  y3 − y 2 
2
 y1 − y3 
2
 y2 − y1 
2

       
  l1   l2   l3  
ε xx   2 2 2  ε n1 
    x2 − x3   x3 − x1   x1 − x2   ε  EQ.5.3.15.6
ε yy  =      
γ   l
 1  l
 2  l
 3   n2 
 xy    y − y  x − x   y − y  x − x   y − y  x − x  γ n3 
2 3 2
 2 3  2 1 3  3 1  2 2 1  1 2 
  l1  l1   l2  l2   l3  l3 
 
Figure 5.3.20 – Neighbor elements for a triangle
5
2
ψ2
β2 γ2
4
ψ3
β3
3
ϕ3
β1 γ1
ϕ1 1
5.3.15.2 Boundary conditions application

As the side rotation of the element is computed using the out-of-plane displacement of the neighbor elements,
the application of clamped or free boundary conditions needs a particular attention. It is natural to consider the
boundary conditions on the edges by introducing a virtual and symmetric element outside of the edge as
described in Figure 5.3.21. In the case of free rotation at the edge, the normal strain ε nk is vanished. From
EQ.5.3.15.4, this leads to:
α k = −α m EQ.5.3.15.7
In EQ.5.3.15.5 the fourth row of the matrix is then changed to:
[cos β 2
h1
cos β 1
h2
− 1
h3
0 0 0 ] EQ.5.3.15.8
01-Jan-2017 71
The clamped condition is introduced by the symmetry in out-of-plane displacement i.e. wm=wk. This implies
α k = α m . The fourth row of the matrix in EQ.5.3.15.5 is then changed to:
[− cos β 2
h1
− cosh2β1 − 1
h3
0 0 0 ] EQ.5.3.15.9
Figure 5.3.21 – Virtual element definition for boundary conditions application
4
5
2
01-Jan-2017 72
5.4 Solid-Shell Elements

Solid-shell elements form a class of finite element models intermediate between thin shell and conventional solid
elements. From geometrical point of view, they are represented by solid meshes with two nodes through the
thickness and generally without rotational degree-of-freedom. On the other hand, they account for shell-like
behavior in the thickness direction. They are useful for modeling shell-like portions of a 3D structure without the
need to connect solid element nodes to shell nodes (Figure 5.4.1).
Figure 5.4.1 Solid-shell elements application
The derivation of solid-shell elements is more complicated than that of standard solid elements since they are
prone to the following problems:
• Shear and membrane locking with the hybrid strain formulation [89], [90], the hybrid stress [91], and
the Assumed and Enhanced Natural Strain formulations [92], [93], [94], and [95].
• Trapezoidal locking caused by deviation of mid-plane from rectangular shape [8].
• Thickness locking due to Poisson’s ratio coupling of the in-plane and transverse normal stresses [89],
[90], [92], and [94].
Solid shell elements in RADIOSS are the solid elements with a treatment of the normal stresses in the thickness
direction. This treatment consists of ensuring constant normal stresses in the thickness by a penalty method.
Advantage of this approach with respect to the plane-stress treatment is that it can simulate the normal
deformability and exhibits no discernible locking problems. The disadvantage is its possible small time step
since it is computed as solid element and the characteristic length is determined often using the thickness.
The solid-shell elements of RADIOSS are the following:

• HA8: 8-node linear solid and solid-shell with or without reduced integration scheme,
• HSEPH: 8-node linear thick shell with reduced integration scheme and physical stabilization of
hourglass modes,
• PA6: Linear pentahedral element for thick shells,
• SHELL16: 16-node quadratic thick shell.
The thick shell elements HA8 and HSEPH are respectively the solid elements HA8 and HEPH in which the
hypothesis of constant normal stress through the thickness is applied by penalty method. The theoretical features
of these elements are explained in section 5.1. The thick shell element SHELL16 is described hereby.
01-Jan-2017 73
5.4.1 Thick Shell Element SHELL16

The element can be used to model thick-walled structures situated between 3D solids and thin shells. The
element is presented in Figure 5.4.2. It has 16 nodes with three translational d.o.f’s per each node. The element is
quadratic in plane and linear through the thickness. The numerical integration through the thickness is carried out
by Gauss-Lobatto schemes rise up to 9 integrations to enhance the quality of elasto-plastic behavior. The in-
plane integration may be done by a reduced 2x2 scheme or a fully integrated 3x3 points (Figure 5.4.3). A
reduced integration method is applied to the normal stress in order to avoid locking problems.
Figure 5.4.2 Thick Shell Element SHELL16
8 15 7
16
5 14 3
13 6
10
1 9 2
The distribution of mass is not homogenous over the nodes. The internal nodes receive three times more mass
than the corner nodes as shown in Figure 5.4.4.
Figure 5.4.3 Integration points for SHELL16
s
r
t
s
t
t
(a) Two to nine integration points (b) In-plane 2x2 and 3x3 integration
through the thickness schemes
01-Jan-2017 74
Figure 5.4.4 Mass distribution for SHELL16 element
3/32 Mo
1/32 Mo
01-Jan-2017 75
5.5 TRUSS ELEMENTS (TYPE 2)

Truss elements are simple two node linear members that only take axial extension or compression. Figure 5.5.1
shows a truss element.
Figure 5.5.1 Truss Element
5.5.1 Property input

The only property required by a truss element is the cross sectional area. This value will change as the element is
deformed. The cross sectional area is computed using:
Area(t − ∆t )
Area(t ) = EQ. 5.5.1.1
(1 − vεɺx ∆t )2
where ν is the Poisson's ratio defined in the material law.
5.5.2 Stability
Determining the stability of truss elements is very simple. The characteristic length is defined as the length of the
element, i.e. the distance between N1 and N2 nodes.
L(t )
∆t ≤ EQ. 5.5.2.1
C
L(t ) is the current truss length and C =

E
Where, is the sound speed.
ρ
5.5.3 Rigid body motion

The rigid body motion of a truss element as shown in Figure 5.5.2 shows the different velocities of nodes 1 and
2. It is the relative velocity difference between the two nodes that produces a strain in the element, namely ex.
Figure 5.5.2 Truss Motion
01-Jan-2017 76
5.5.4 Strain
The strain rate, as shown in Figure 5.5.2, is defined as:
εɺx = =
(
∂v x ∂ V ⋅ ex ) EQ. 5.5.4.1
∂x ∂x
5.5.5 Material type

A truss element may only be assigned two types of material properties. These are types 1 and 2, elastic and
elasto-plastic properties respectively.
5.5.6 Force calculation

The calculation of forces in a truss element is performed by explicit time integration:
F (t + ∆t ) = F (t ) + Fɺ∆t EQ. 5.5.6.1

A generalized force-strain graph can be seen in Figure 5.5.3. The force rate under elastic deformation is given
by:
Fɺx = EAεɺx EQ. 5.5.6.2
where:
E is the Elastic Modulus,
A is the cross sectional area.
In the plastic region, the force rate is given by:
Fɺx = Et Aεɺx EQ. 5.5.6.3
where Et is the gradient of the material curve at the deformation point.
Figure 5.5.3 Force-Strain Relationship
(a) without gap (b) with gap
l
log
gap l0
In a general case, it is possible to introduce a gap distance in the truss definition. If gap is not null, the truss is
activated when the length of the element is equal to the initial length minus the gap value. This results a force-
strain curve shown in Figure 5.5.3(b).
01-Jan-2017 77
5.6 BEAM ELEMENTS (TYPE 3)

RADIOSS uses a shear beam theory or Timoshenko formulation for its beam elements.
This formulation assumes that the internal virtual work rate is associated with the axial, torsional and shear
strains. The other assumptions are:
• No cross section deformation in its plane.
• No cross section warping out of its plane.
With these assumptions, transverse shear is taken into account.
This formulation can degenerate into a standard Euler-Bernoulli formulation (the cross section remains normal to
the beam axis). This choice is under user control.
5.6.1 Local coordinate system

The properties describing a beam element are all defined in a local coordinate system.
This coordinate system can be seen in Figure 5.6.1. Nodes 1 and 2 of the element are used to define the local X
axis, with the origin at node 1. The local Y axis is defined using node 3, which lies in the local XY plane, along
with nodes 1 and 2. The Z axis is determined from the vector cross product of the positive X and Y axes.
The local Y direction is first defined at time t=0 and its position is corrected at each cycle, taking into account
the mean rotation of the X axis. The Z axis is always orthogonal to the X and Y axes.
Deformations are computed with respect to the local coordinate system displaced and rotated to take into account
rigid body motion. Translational velocities V and angular velocities Ω with respect to the global reference
frame are expressed in the local frame.
Figure 5.6.1 Beam Element Local Axis
5.6.2 Beam element geometry

The beam geometry is user-defined by:
A: cross section area,
Ix: moment of inertia of cross section about local x axis,
Iy: moment of inertia of cross section about local y axis,
Iz: moment of inertia of cross section about local z axis.
01-Jan-2017 78
The moments of inertia about the y and z axes are concerned with bending. They can be calculated using the
relationships:
I y = ∫∫ z 2 dydz EQ. 5.6.2.1

A
I z = ∫∫ y 2 dydz EQ. 5.6.2.2

A
The moment of inertia about the x axis concerns torsion. This is simply the summation of the previous two
moments of Ontario:
Ix = Iy + Iz EQ. 5.6.2.3
5.6.3 Minimum time step

The minimum time step for a beam element is determined using the following relation:
aL
∆t = EQ. 5.6.3.1
c
where:
c is the speed of sound : E/ρ ,
a = 0.5 min(4,1 + b / 12) F1 b / 3 F2 ,
F1 = (1 + 2d ) − d
2
2

( 2

)
F2 = min  F1 , 1 + 2d s  − d s 2
 12 12 * E 
d s = d * max1, * 1+ * b * (1 − Ishear ) 
 b 5/ 6*G 
d = max(d m , d f )
AL2
b=
max (I y , I z )
5.6.4 Beam element behavior

RADIOSS beam elements behave in four individual ways:
• Membrane or axial deformation.
• Torsion.
• Bending about the z axis.
• Bending about the y axis.
5.6.4.1 Membrane behavior

Membrane or axial behavior is the extension or compression of the beam element. The forces acting on an
element are shown in Figure 5.6.2.
01-Jan-2017 79
Figure 5.6.2 Membrane Forces
The force rate vector for an element is calculated using the relation:
 Fɺx1  EA + 1 − 1 υ x1 
ɺ =
l  − 1 + 1 υ x 2 
EQ. 5.6.4.1
 Fx 2 
where:
E is the elastic modulus,
l is the beam element length,
υx is the nodal velocity in x direction.
With the force rate equation, the force vector is determined using explicit time integration:
Fx (t + ∆t ) = Fx (t ) + Fɺx ∆t EQ. 5.6.4.2
5.6.4.2 Torsion
Torsional deformation occurs when the beam is loaded with a moment M about the X axis as shown in Figure
5.6.3.
Figure 5.6.3 Torsional Loading
The moment rate vector is computed by:
 Mɺ x1  GI x + 1 − 1 θɺx1 
 ɺ =  
l  − 1 + 1 θɺx 2 
EQ. 5.6.4.3
M x 2 
where:
G is the modulus of rigidity,
θɺx is the angular rotation rate.
The moment about the X axis is found by explicit time integration:
M x (t + ∆t ) = M x (t ) + Mɺ x ∆t EQ. 5.6.4.4
5.6.4.3 Bending about z axis

Bending about the z axis involves a force in the y direction and a moment about the z axis as shown in Figure
5.6.4.
01-Jan-2017 80
Figure 5.6.4 Bending about the z axis
Two vector fields must be solved for forces and moments. The rate equations are:
 v y1 
 
 F y1 
ɺ EI z  + 12 6l − 12 + 6l  θɺz1 
ɺ = 3
 Fy 2  l (1 + φ y )  − 12 − 6l + 12 − 6l  v y 2 
  EQ. 5.6.4.5
ɺ 
θ z 2 
 v y1 
 Mɺ z1  EI z + 6l (4 + φ )l 2
− 6l (2 − φ y )l  θɺz1 
2
 ɺ = 3
y
 M z 2  l (1 + φ y ) + 6l

(2 − φ )l
y
2
− 6l (4 + φ y )l 2  v y 2 
 EQ. 5.6.4.6
ɺ 
θ z 2 
144(1 + v )I z
where φy = ,
5 Al 2
ν is the Poisson's Ratio.
The factor Φ y takes into account transverse shear.

The time integration for both is:
Fy (t + ∆t ) = Fy (t ) + Fɺy ∆t EQ. 5.6.4.7
M z (t + ∆t ) = M z (t ) + Mɺ z ∆t EQ. 5.6.4.8
5.6.4.4 Bending about Y axis

Bending about the Y axis is identical to bending about the Z axis. A force in the Y direction and a moment about
the Z axis, shown in Figure 5.6.5, contribute to the elemental bending.
Figure 5.6.5 Bending about Y axis
01-Jan-2017 81
The rate equations are:
ν z1 
 
 Fɺz1  El y + 12 6l − 12 + 6l  θɺy1 
ɺ = 3   EQ. 5.6.4.9
 Fz 2  l (1 + Φ z )  − 12 − 6l + 12 − 6l  ν z 2 
ɺ 
θ y 2 
ν z1 
 
 Mɺ y1  El y + 6l (4 + Φ z )l 2 − 6l (2 − Φ z )l 2  θɺy1 
 ɺ = 3  2
EQ. 5.6.4.10
 M y 2  l (1 + Φ z ) + 6l (2 − Φ z )l − 6l (4 + Φ z )l  ν z 2 
2
θ y 2 
ɺ
where:
144(1 +ν )I y
Φz = .
5 Al 2
Like bending about the Z axis, the factor Φ z introduces transverse shear.
With the time integration, the expression is:
Fz (t + ∆t ) = Fz (t ) + Fɺz ∆t EQ. 5.6.4.11
M y (t + ∆t ) = M y (t ) + Mɺ y ∆t EQ. 5.6.4.12
5.6.5 Material properties

A beam element may have two different types of material property:
• Elastic
• Elasto-plastic
5.6.5.1 Elastic Behavior

The elastic beam is defined using material law 1 which is a simple linear material law.
The cross-section of a beam is defined by its area A and three moments of inertia Ix , Iy and Iz .
An elastic beam can be defined with these four parameters. For accuracy and stability, the following limitations
should be respected:
L> A EQ. 5.6.5.1
0.01A2 < I y < 100 A2 EQ. 5.6.5.2
0.01A2 < I z < 100 A2 EQ. 5.6.5.3
0.1(I y + I z ) < I x < 10(I y + I z )

EQ. 5.6.5.4
01-Jan-2017 82
5.6.5.2 Elasto-plastic Behavior

A global plasticity model is used.
The main assumption is that the beam cross section is full and rectangular. Optimal relations between sections
and section inertia are:
12 I y I z = A4 EQ. 5.6.5.5
Ix = Iy + Iz EQ. 5.6.5.6
However, this model also gives good results for the circular or ellipsoidal cross-section. For tubular or H cross-
sections, plasticity will be approximated.
Recommendations:
L> A EQ. 5.6.5.7
0.1A4 < 12 I y I z < 10 A4 EQ. 5.6.5.8
0.01 < I y / I z < 100 EQ. 5.6.5.9
0.5(I y + I z ) < I x < 2(I y + I z ) EQ. 5.6.5.10
5.6.5.3 Global Beam Plasticity

The elasto-plastic beam element is defined using material law 2:
εɺ 
σ y = (A + Bε pn )1 + C ln

 EQ. 5.6.5.11
 εɺ0 
The increment of plastic strain is:
∆W plastic
∆ε p = EQ. 5.6.5.12
σy
The equivalent strain rate is derived from the total energy rate:
∆Wtotal
εɺ = EQ. 5.6.5.13
σ eq ∆t
Yield stress:
Fx2 3  M x2 M y M z2 
2
σ eq = + + + EQ. 5.6.5.14
A2 A  I xx I yy I zz 
If σ eq > σ y , one performs a radial return on the yield surface:

σy
Fxpa = Fx EQ. 5.6.5.15
σ eq
and for i= x, y, z:
σy
M ipa = M i EQ. 5.6.5.16
σ eq
01-Jan-2017 83
5.6.6 Inertia Computation

The computational method of inertia for some kinds of elements as beam is particular as the inertia has to be
transferred to the extremities of the beam. The nodal inertias are computed in function of the material density
ρ , the cross-section area S, the element length L and the moments of inertia I xx , I yy , I zz :
  ρSL  L2   ρL   ρL  
MAX     +   • MAX (I yy ; I zz ) ;   • I xx  EQ. 5.6.6.1
  2  12   2   2  
01-Jan-2017 84
5.7 ONE DEGREE OF FREEDOM SPRING ELEMENTS

(TYPE 4)
One degree of freedom (DOF) spring elements are defined as a type 4 property set. Three variations of the
element are possible:
• Spring only
• Dashpot (damper) only
• Spring and dashpot in parallel
These three configurations are shown in Figures 5.7.1 to 5.7.3.
Figure 5.7.1 Spring Only
Figure 5.7.2 Dashpot Only
Figure 5.7.3 Spring and Dashpot in Parallel
No material data card is required for spring elements. However, the stiffness k and equivalent viscous damping
coefficient c are required. The mass m is required if there is any spring translation.
There are three other options defining the type of spring stiffness with the hardening flag:
• Linear Stiffness
• Nonlinear Stiffness
• Nonlinear Elasto-Plastic Stiffness
01-Jan-2017 85
Likewise, the damping can be either:

• Linear
• Nonlinear
A spring may also have zero length. However, a one DOF spring must have 2 nodes.
The forces applied on the nodes of a one DOF spring are always colinear with direction through both nodes;
refer to figure 5.7.4.
Figure 5.7.4 Colinear Forces
5.7.1 Time step

The time of a spring element depends on the values of stiffness, damping and mass.
For a spring only element:
m
∆t = EQ. 5.7.1.1
k
For a dashpot only element:
m
∆t = EQ. 5.7.1.2
2c
For a parallel spring and dashpot element:
∆t =
( mk + c )− c2
EQ. 5.7.1.3
k
The critical time step ensures that the stability of the explicit time integration is maintained, but it does not
ensure high accuracy of spring vibration behavior. Only two time steps are required during one vibration period
of a free spring to keep stability. However, if true sinusoidal reproduction is desired, the time step should be
reduced by a factor of at least 5.
If the spring is used to connect the two parts, the spring vibration period increases and the default spring time
step ensures stability and accuracy.
5.7.2 Linear spring

Function number defining f(δ)
N1=0
01-Jan-2017 86
The general linear spring is defined by constant mass, stiffness and damping. These are all required in the
property type definition. The relationship between force and spring displacement is given by:
F = k (l − l0 ) + c
dl
EQ. 5.7.2.1
dt
Figure 5.7.5 Linear Force-Displacement Curve
The stability condition is given by equation 5.7.1.3.:
∆t =
(c 2
)
+ km − c
EQ. 5.7.2.2
k
5.7.3 Nonlinear elastic spring

Hardening flag
H=0
The hardening flag must be set to 0 for a nonlinear elastic spring. The only difference between linear and
nonlinear elastic spring elements is the stiffness definition. The mass and damping are defined as constant.
However, a function must be defined that relates the force, F, to the displacement of the spring, (l-l0). It is
defined as:
F = f (l − l0 ) + c
dl
EQ. 5.7.3.1
dt
01-Jan-2017 87
Figure 5.7.6 Nonlinear Elastic Force-Displacement Curve
The stability criterion is the same as for the linear spring, but rather than being constant, the stiffness is
displacement dependent:
∆t =
(c 2
)
+ k ′m − c
EQ. 5.7.3.2
k′
where:
 ∂ 
k ′ = max  f (l − l0 ) EQ. 5.7.3.3
 ∂ (l − l0 ) 
5.7.4 Nonlinear elasto-plastic spring - Isotropic hardening

H=1
The hardening flag must be set to 1 in this case and f(l-l o ) is defined by a function. Hardening is isotropic if
compression behavior is identical to tensile behavior:
F = f (l − l0 ) + C
dl
EQ. 5.7.4.1
dt
Figure 5.7.7 Isotropic Hardening Force-Displacement Curve
01-Jan-2017 88
5.7.5 Nonlinear elasto-plastic spring - Decoupled hardening

H=2
The hardening flag is set to 2 in this case and f(l-l o ) is defined by a function. The hardening is decoupled for
compression and tensile behavior:
F = f (l − l0 ) + C
dl
EQ. 5.7.5.1
dt
Figure 5.7.8 Decoupled Hardening Force-Displacement Curve
5.7.6 Nonlinear elastic-plastic spring - Kinematic hardening

H=4
The hardening flag is set to 4 in this case and f1(l-l o ) and f2(l-l o ) (respectively maximum and minimum yield
force) are defined by a function. The hardening is kinematic if maximum and minimum yield curves are
identical:
F = f (l − l0 ) + C
dl
EQ. 5.7.6.1
dt
Figure 5.7.9 Kinematic Hardening Force-Displacement Curve
01-Jan-2017 89
5.7.7 Nonlinear elasto-plastic spring - Nonlinear unloading

H=5
The hardening flag is set to 5 in this case and f(δ) and f2(δmax) (respectively maximum yield force and residual
deformation) are defined by a function. Uncoupled hardening in compression and tensile behavior with nonlinear
unloading:
F = f (l − l0 ) + C
dl
EQ. 5.7.7.1
dt
With δ= l-l o .
Figure 5.7.10 Nonlinear unloading Force-Displacement Curve
5.7.8 Nonlinear dashpot

The input properties for a nonlinear dashpot are very close to that of a spring. The required values are:
• Mass, M.
• A function defining the change in force with respect to the spring displacement. This must be equal
to unity:
f (l − l0 ) = 1
• A function defining the change in force with spring displacement rate,
g (dl / dt )
• The hardening flag in the input must be set to zero.
The relationship between force and spring displacement and displacement rate is:
 dl   dl 
F = f (l − l0 )g   = g   EQ. 5.7.8.1
 dt   dt 
A nonlinear dashpot property is shown in Figure 5.7.11.
01-Jan-2017 90
Figure 5.7.11 Nonlinear Dashpot Force Curve
The stability condition for a nonlinear dashpot is given by:
M
∆t = EQ. 5.7.8.2
C′
where:
 ∂ 
C ′ = max  g (dl / dt ) EQ. 5.7.8.3
 ∂(dl / dt ) 
5.7.9 Nonlinear viscoelastic spring

The input properties for a nonlinear viscoelastic spring are:
• Mass, M
• Equivalent viscous damping coefficient C
• A function defining the change in force with spring displacement
f (l − l0 )
• A function defining the change in force with spring displacement rate
g (dl / dt )
The hardening flag in the input must be set to equal zero. The force relationship is given by:
 dl 
F = f (l − l0 )g   EQ. 5.7.9.1
 dt 
Graphs of this relationship for various values of g (dl / dt ) are shown in Figure 5.7.12.
01-Jan-2017 91
Figure 5.7.12 Visco-Elastic Spring Force-Displacement Curves
The stability condition is given by:
∆t =
( C′ + k ′M )− C′
2
EQ. 5.7.9.2
k′
where:
 ∂ 
K ′ = max  f (l − l0 ) EQ. 5.7.9.3
 ∂ (l − l0 ) 
 ∂ 
C ′ = max  g (dl / dt )
 ∂(dl / dt )
EQ. 5.7.9.4

01-Jan-2017 92
5.8 GENERAL SPRING ELEMENTS (TYPE 8)

General spring elements are defined as type 8 element property. They are mathematical elements, which have 6
DOF, three translational displacements and three rotational degrees of freedom. Each DOF is completely
independent from the others. Spring displacements refer to either spring extension or compression. The stiffness
is associated to each DOF. Directions can either be global or local. Local directions are defined with a fixed or
moving skew frame. Global force equilibrium is respected, but without global moment equilibrium. Therefore,
this type of spring is connected to the laboratory that applies the missing moments, unless the two defining nodes
are not coincident.
5.8.1 Time step

The time step calculation for general spring elements is the same as the calculation of the equivalent type 4
spring (Section 5.7.1).
5.8.2 Linear spring

See section 5.7.2; the explanation is the same as for spring type 4.



5.8.6 Nonlinear elasto-plastic spring - Kinematic hardening



5.8.9 Nonlinear viscoelastic spring

5.8.10 Skew frame properties

To help understand the use of skew frames, the deformation in the local x direction of the spring will be
considered. If the skew frame is fixed, deformation in the local X direction is shown in Figure 5.8.1:
01-Jan-2017 93
Figure 5.8.1 Fixed Skew Frame
The same local x direction deformation, with a moving skew frame, can be seen in Figure 5.8.2.
Figure 5.8.2 Moving Skew Frame
In both cases, the forces are in equilibrium, but the moments are not. If the first two nodes defining the moving
skew system are the same nodes as the two spring element nodes, the behavior becomes exactly the same as that
of a type 4 spring element. In this case the momentum equilibrium is respected and local Y and Z deformations
are always equal to zero.
Fixed Nodes
If one of the two nodes is completely fixed, the momentum equilibrium problem disappears. For example, if
node 1 is fixed, the force computation at node 2 is not dependent on the location of node 1. The spring then
becomes a spring between node 1 and the laboratory, as shown in Figure 5.8.3.
Figure 5.8.3 Fixed node - Fixed skew frame
01-Jan-2017 94
It is generally recommended that a general spring element (type 8) be used only if one node is fixed in all
directions or if the two nodes are coincident. If the two nodes are coincident, the translational stiffness’ have to
be large enough to ensure that the nodes remain near coincident during the simulation.
5.8.11 Deformation sign convention

Positive and negative spring deformations are not defined with the variation of initial length. The initial length
can be equal to zero for all or a given direction. Therefore, it is not possible to define the deformation sign with
length variation.
The sign convention used is the following. A deformation is positive if displacement (or rotation) of node 2
minus the displacement of node 1 is positive. The same sign convention is used for all 6 degrees of freedom.
ui = ui 2 − ui1 EQ. 5.8.11.1
θ i = θ i 2 − θ i1 EQ. 5.8.11.2
5.8.12 Translational forces

The translational forces that can be applied to a general spring element can be seen in Figure 5.8.4. For each
DOF (i = x, y, z), the force is calculated by:
Fi = f i (ui ) + Ci uɺi EQ. 5.8.12.1
where:
C is the equivalent viscous damping coefficient
f i (ui ) is a force function related to spring displacement

The value of the displacement function depends on the type of general spring being modeled.
Figure 5.8.4 Translational Forces
5.8.12.1 Linear Spring

If a linear general spring is being modeled, the translation forces are given by:
Fi = K iui + Ci uɺi EQ. 5.8.12.2
where K is the stiffness or unloading stiffness (for elasto-plastic spring)
5.8.12.2 Nonlinear Spring

If a nonlinear general spring is being modeled, the translation forces are given by:
01-Jan-2017 95
  uɺ  
Fi = f i (ui ) A + B ln i  + g (uɺi ) + Ci uɺi EQ. 5.8.12.3
 D 
where:
f (ui ) is a function defining the change in force with spring displacement
g (uɺi ) is a function defining the change in force with spring displacement rate
A = coefficient. Default = 1.
B = coefficient
D = coefficient. Default = 1.
5.8.13 Moments
Moments can be applied to a general spring element, as shown in Figure 5.8.5. For each DOF (i = x, y, z), the
moment is calculated by:
M i = f i (θ i ) + Ciθɺi EQ. 5.8.13.1
where:
C is the equivalent viscous damping coefficient
f i (θ i ) is a force function related to spring rotation

The value of the rotation function depends on the type of general spring being modeled. Not all functions and
coefficients defining moments and rotations are of the same value as that used in the translational force
calculation.
Figure 5.8.5 General Spring Moments
5.8.13.1 Linear Spring

If a linear general spring is being modeled, the moments are given by:
M i = K iθ i + Ciθɺi EQ. 5.8.13.2
where K is the stiffness or unloading stiffness (for elasto-plastic spring).
01-Jan-2017 96
5.8.13.2 Nonlinear Spring

If a nonlinear general spring is being modeled, the moments are given by:
  θɺ  
M i = f (θ i ) A + B ln i
 
( ) + C θɺ
 + g θɺi i i EQ. 5.8.13.3
 D  
where:
f (θ i ) is a function defining the change in force with spring displacement
g (θ i ) is a function defining the change in force with spring displacement rate

A = coefficient. Default = 1.
B = coefficient.
D = coefficient. Default = 1.
5.8.14 Multidirectional failure criteria

Flag for rupture criteria: Ifail
Ifail=1
The rupture criteria flag is set to 1 in this case:
F 2 = Dx2 + D y2 + Dz2 + Dxx2 + D yy2 + Dzz2 EQ. 5.8.14.1
Where:
Dx = Dxp is the rupture displacement in positive x direction if u x > 0
Dx = Dxn is the rupture displacement in negative x direction if u x > 0

Graphs of this rupture criterion can be seen in Figure 5.8.6.
Figure 5.8.6 Multi-directional failure criteria curves
Dyp
Dxn Dxp
Dyn
01-Jan-2017 97
5.9 PULLEY TYPE SPRING ELEMENTS (TYPE12)

Pulley type springs are defined by type 12 element property. A general representation can be seen in Figure
5.9.1. It is defined with three nodes, where node 2 is located at the pulley position. Other properties such as
stiffness, damping, nonlinear and plastic effects are the same as for the other spring types, and are defined using
the same format.
A deformable "rope" joins the three nodes, with the mass distribution as follows: one quarter at node 1; one
quarter at node 3 and one half at node 2.
Coulomb friction can be applied at node 2, which may also take into account the angle between the two rope
strands.
The two rope strands have to be long enough to avoid node 1 or node 3 sliding up to node 2 (the pulley). If this
occurs, either node 1 or 3 will be stopped at node 2, just as if there were a knot at the end of the rope.
Figure 5.9.1 Pulley Type Spring Element Representation
5.9.1 Time step

The time step is calculated using the relation:
∆t =
( 2KM + C )− C2
EQ. 5.9.1.1
2K
This is the same as for type 4 spring elements, except that the stiffness is replaced with twice the stiffness to
ensure stability with high friction coefficients.
5.9.2 Linear spring



01-Jan-2017 98


5.9.7 Nonlinear visco-elastic spring

5.9.8 Friction effects

Pulley type springs can be modeled with or without Coulomb friction effects.
5.9.8.1 Without Friction

Without friction, the forces are computed using:
dδ
F1 = F2 = Kδ + C EQ. 5.9.8.1
dt
where:
δ is the total rope elongation = l − l0 with l = l12 + l23

K is the rope stiffness
C is the rope equivalent viscous damping
5.9.8.2 With Coulomb Friction

If Coulomb friction is used, forces are corrected using:
 dδ dδ 
F = K (δ1 + δ 2 ) + C  1 + 2  EQ. 5.9.8.2
 dt dt 
  dδ dδ   βµ  
∆F = max K (δ1 + δ 2 ) + C  1 + 2 , F tanh    EQ. 5.9.8.3
  dt dt   2 
F1 = F + ∆F EQ. 5.9.8.4
01-Jan-2017 99
F3 = F − ∆F EQ. 5.9.8.5
F2 = − F1 − F3 EQ. 5.9.8.6
where:
δ1 is the elongation of strand 1-2
δ2 is the elongation of strand 2-3
01-Jan-2017 100
5.10 BEAM TYPE SPRING ELEMENTS (TYPE13)

Beam type spring elements are defined as property type 13 elements. This type of spring element functions as if
it were a beam element. The six independent modes of deformation are:
• Traction / compression
• Torsion
• Bending (two modes)
• Shear (two modes)
Beam type springs only function if their length is not zero. A physical representation of a beam type spring can
be seen in Figure 5.10.1.
Figure 5.10.1 Representation of Beam Type Spring
5.10.1 Time step

5.10.1.1 Translational stiffness time step
mass. max(Kt ) + C t2 − Ct
∆t translational _ stifness =
max(K t )
Where, max(Kt) is the maximum translational stiffness
Ct is the translational damping
5.10.1.2 Rotational stiffness time step

inertia.K r' + C ' r2 − C ' r
∆t rotational _ stifness =
K r'
Kr’ is the equivalent rotational stiffness: K r = max( K t ). L + max( K r )
' 2
Where, max(Kt) is the maximum translational stiffness

max(Kr) is the maximum rotational stiffness
Cr’ is the equivalent rotational damping: C r = max(C t ). L + max(C r )
' 2
Where,
max(Ct) is the maximum translational damping
max(Cr) is the maximum rotational damping
01-Jan-2017 101
5.10.2 Linear spring

The properties required to define the spring characteristics are stiffness K and damping C. Nonlinear and elasto-
plastic properties can also be applied, for all degrees of freedom. The properties are of the same form as simple
type 4 spring elements (section 5.7).



5.10.6 Nonlinear elasto-plastic spring - Kinematic hardening



5.10.9 Nonlinear visco-elastic spring

5.10.10 Skew frame properties

Beam type spring elements are best defined using three nodes (Figure 5.10.2). Nodes 1 and 2 are the two ends of
the element and define the local X axis. Node 3 allows the local Y and Z axes to be defined. However, this node
does not need to be supplied.
If all three nodes are defined, the local reference frame is calculated by:
x = n1n2 EQ. 5.10.10.1
z = x × n1n3 EQ. 5.10.10.2
y = z×x EQ. 5.10.10.3

If node 3 is not defined, the local skew frame that can be specified for the element is used to define the Z axis.
The X and Y axes are defined in the same manner as before.
z = x × yskew EQ. 5.10.10.4
01-Jan-2017 102
If no skew frame and no third node are defined, the global Y axis is used to replace the Y skew axis. If the Y
skew axis is collinear with the local X axis, the local Y and Z axes are placed in a totally arbitrary position. The
local Y axis is defined at time zero, and is corrected at each cycle, taking into account the mean X axis rotation.
Figure 5.10.2 Element Definition
5.10.11 Sign Conventions

The sign convention used for defining positive displacements and forces can be seen in Figure 5.10.3.
Figure 5.10.3 Sign Conventions
5.10.12 Tension
The tension component of the beam type spring element is independent of other forces. It is shown in Figure
5.10.4. The tension at each node is computed by:
Fx1 = f x (u x ) + C xuɺ x EQ. 5.10.12.1
Fx 2 = − Fx1 EQ. 5.10.12.2
01-Jan-2017 103
where
u x = u x 2 − u x1 is the relative displacement of nodes 1 and 2.
f x (u x ) is the function defining the force-displacement relationship.

It can be linear or nonlinear (see sections 5.7.2 to 5.7.5.).
Figure 5.10.4 Spring Tension
5.10.13 Shear - XY
Shear in the Y direction along the face perpendicular to the X axis is a combination of forces and moments. This
can be seen in Figure 5.10.5.
Figure 5.10.5 XY Shear Forces and Moments
There are two mechanisms that can cause shear. The first is the beam double bending as shown in Figure 5.10.5.
The second is shear generated by node displacement, as shown in Figure 5.10.6, where node 2 is displaced.
Figure 5.10.6 Shear due to Node Displacement
01-Jan-2017 104
The forces and moments are calculated by:
Fy1 = f y (u y ) + C y uɺ y EQ. 5.10.13.1
Fy 2 = − Fy1 EQ. 5.10.13.2
M z1 = −lFy 2 EQ. 5.10.13.3
M z 2 = M z1 EQ. 5.10.13.4
where:
 θ + θ z1 
u y = u y 2 − u y1 − l  z 2 
 2 
f y (u y ) is the function defining the force-displacement relationship.
5.10.14 Shear - XZ
The XZ shear is orthogonal to the XY shear described in the previous section. The forces and moments causing
the shear can be seen in Figure 5.10.7.
Figure 5.10.7 XZ Shear Forces and Moments
The forces and moments are calculated by:
Fz1 = f z (u z ) + C z uɺ z EQ. 5.10.14.1
Fz 2 = − Fz1 EQ. 5.10.14.2
M y1 = lFz 2 EQ. 5.10.14.3
M y 2 = M y1 EQ. 5.10.14.4
Where,
 θ y 2 + θ y1 
u z = u z 2 − u z1 + l  
 2 
f z (u z ) is the function defining the force-displacement relationship.
01-Jan-2017 105
5.10.15 Torsion
Torsional forces, shown in Figure 5.10.8, are calculated using the relations:
M x1 = f xx (θ x ) + C xxθɺx EQ. 5.10.15.1
M x 2 = − M x1 EQ. 5.10.15.2
where:
θ x = θ x 2 − θ x1 is the relative rotation of node 1 and 2.
f xx (θ x ) is the function defining the force-displacement relationship.
Figure 5.10.8 Beam Type Spring Torsion
5.10.16 Bending about the Y Axis

Bending about the Y axis can be seen in Figure 5.10.9. The equations relating to the moments being produced
are calculated by:
M y1 = f yy (θ y ) + C yyθɺy EQ. 5.10.16.1
M y 2 = − M y1 EQ. 5.10.16.2
where:
θ y = θ y 2 − θ y1 is the relative rotation of node 1 and 2.
f yy (θ y ) is the function defining the force-displacement relationship.
Figure 5.10.9 Bending about Y axis
5.10.17 Bending about the Z Axis

The equations relating to the moment generated in a beam type spring element and the beam's displacement,
(Figure 5.10.10) is given by:
M z1 = f zz (θ z ) + C zzθɺz EQ. 5.10.17.1
M z 2 = − M z1 EQ. 5.10.17.2
01-Jan-2017 106
where θ z = θ z 2 − θ z1 is the relative rotation of node 1 and 2.
f zz (θ z ) is the function defining the force-displacement relationship.

Figure 5.10.10 Bending about Z axis
5.10.18 Multidirectional failure criteria

See Section 5.8.14; the explanation is the same as for spring type 8.
01-Jan-2017 107
5.11 MULTISTRAND ELEMENTS (TYPE 28)
5.11.1 Introducing Multistrand Elements

Multistrand elements are n-node springs where matter is assumed to slide through the nodes. It could be used for
belt modelization by taking nodes upon the dummy. Friction may be defined at all or some nodes. When nodes
are taken upon a dummy in order to modelize a belt, this allows friction to be modelized between the belt and the
dummy.
5.11.2 Internal Forces Computation
Nodes are numbered from 1 to n, and strands are numbered from 1 to n-1 (strand k goes from node Nk to node
Nk+1).
5.11.2.1 Averaged force into multistrand element

The averaged force in the multistrand is computed as:
K C
Linear spring F= 0
δ + 0 δɺ
L L
F = f (ε ) ⋅ g (εɺ ) +
C ɺ
Non linear spring δ
L0
F = f (ε ) +
C ɺ
or δ if g function identifier is 0
L0
F = g (εɺ ) +
C ɺ
or δ if f function identifier is 0
L0
L − L0
where, ε is engineering strain: ε=
L0
L0 is the reference length of element.
5.11.2.2 Force into each strand

The force into each strand k is computed as:
Fk=F+∆Fk
01-Jan-2017 108
∆Fk is computed an incremental way:
∆Fk (t ) = ∆Fk (t − 1) +
K K
0
δε k − 0 δε EQ. 5.11.2.1
lk L
with lk0 the length of the unconstrained strand k, δε = ε (t ) − ε (t − 1) and δε k = δtu k ⋅ (vk +1 − vk ) ,
where u k is the unitary vector from node Nk to node Nk+1 .
Assuming:
lk L
0
= 0 EQ. 5.11.2.2
lk L
where lk is the actual length of strand k.
Therefore, EQ. 5.11.2.1 reduces to:
K 
∆Fk (t ) = ∆Fk (t − 1) +
L
 δε k − δε 
0 
EQ. 5.11.2.3
l  lk 
5.11.2.3 Friction
Friction is expressed at the nodes: if µ is the friction coefficient at node k, the pulley friction at node Nk is
expressed as:
 βµ 
∆Fk −1 − ∆Fk ≤ (2 F + ∆Fk −1 + ∆Fk ) tanh  EQ. 5.11.2.4
 2 
 βµ 
When equation 5.11.2.4 is not satisfied, ∆Fk −1 − ∆Fk is reset to (2 F + ∆Fk −1 + ∆Fk ) tanh  .
 2 
All the ∆Fk (k=1,n-1) are modified in order to satisfy all conditions upon ∆Fk −1 − ∆Fk (k=2,n-1), plus the
following condition on the force integral along the multistrand element:
∑ l (F + ∆F ) = LF
k =1, n −1
k k EQ. 5.11.2.5
This process could fail to satisfy equation 5.11.2.4 after the ∆Fk (k = 1, n − 1) modification, since no iteration
is made. However, in such a case one would expect the friction condition to be satisfied after a few time steps.
Note: Friction expressed upon strands (giving a friction coefficient µ along strand k) is related to pulley
friction by adding a friction coefficient µ /2 upon each nodes Nk and Nk+1 .
5.11.2.4 Time step

Stability of a multistrand element is expressed as:
Ck2 + ρlk K k − Ck
∆t ≤ , ∀k EQ. 5.11.2.6
Kk
01-Jan-2017 109
Mass of the multistrand

with Kk = and (assuming 9.2.2.2) :
L0
K F   KL FL 
K k = max 0 ,  = max 0 , 
0 
 lk lk − l k  ( 0 
 lk L lk L − L  ) EQ. 5.11.2.7
 
 f (ε ) (εɺ ) + C 
dg
dεɺ  =  f (ε ) dg (εɺ ) + C  L
Ck =    0 EQ. 5.11.2.8
lk0
 dε  lk L
01-Jan-2017 110
5.12 SPRING TYPE PRETENSIONER (TYPE 32)
5.12.1 Introducing pretensioners

Pretensioner expected behavior is as follows: before pretensioning, a piston is fixed in its initial position; when
activated, the piston is pushed and cannot slide once the piston has reached the end of its slide, it is unable to
slide further in any direction in the opposite direction from its actual position.
5.12.2 RADIOSS model for pretensioners

5.12.2.1 Linear model
STIF0 is the spring stiffness before sensor activation. At sensor activation, the 2 input coefficients among D1,
STIF1, F1 and E1 determine the pretensioner characteristics. Let us recall the following relations between the 4
coefficients:
D1 ⋅ F1 F
E1 = , K1 = 1 EQ. 5.12.2.1
2 D1
STIF0 is also used as unloading stiffness before the end of the piston's slide, and as both loading and unloading
stiffness at the end of the piston's slide. STIF0 should be large enough to allow locking.
5.12.2.2 Nonlinear model
01-Jan-2017 111
STIF0 is spring stiffness before sensor activation. Depending on the input, pretensioning force is defined as f(L-
L0), with either g(t-t0), or f(L-L0).g(t-t0), with L0 length of the spring at sensor activation time and at t0 sensor
activation time.
Similar use of STIF0 allows piston locking.
5.12.2.3 Force computation

Let the pretensioning force Fp (t ) = F1 + STIF1 ⋅ (L(t ) − L0 ) for a linear model,
and Fp (t ) = f (L(t ) − L0 ) or g (t − t 0 ) or f (L (t ) − L0 ) ⋅ g (t − t 0 ) for a nonlinear model.

The force into the pretensioner spring is computed as: EQ. 5.12.2.2
if, Fp (t + dt ) ≥ 0
F (t + dt ) = Max (Fp (t + dt ), F (t ) + STIF0 ⋅ L(t + dt ) − L(t ))
and F (t + dt ) = F (t ) + STIF0 ⋅ (L(t + dt ) − L(t )) otherwise.
01-Jan-2017 112

RADIOSS Element

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RADIOSS Element

Uploaded by

Copyright:

Available Formats

RADIOSS THEORY MANUAL

Version 2017 – January 2017

5.6.3 MINIMUM TIME STEP 79

5.11 MULTISTRAND ELEMENTS (TYPE 28) 108

5.0 ELEMENT LIBRARY

• Solid elements : 8- and 20-node bricks, 4- and 10-node tetrahedrons,

5.1 Solid Hexahedron Elements

5.1.1 Shape functions for linear bricks

Figure 5.1.1 8 Node Brick Element

where the viI are the nodal velocities.

5.1.2 Strain rate

where Fξ is the Jacobian matrix.

Relating the element velocity field to its shape function gives:

5.1.3 Assumed strain rate

The derivation of the shape functions is given by:

{εɺ} = ∑[BI ]{vI } = ∑([BI ]0 +[BI ]H ){vI } = {εɺ}0 +{εɺ}H

Belvtschko and Bindeman [65] ASQBI assumed strain is used:

{εɺ} = ∑([BI ]0 +[BI ]H ){vI }

 X1234I −νYI3 −νYI24 −νZI2 −νZI34

∂φ1 ∂φ3 ∂φ1 ∂φ3

5.1.3.1 Incompressible or quasi-incompressible cases

5.1.4 Internal force calculation

5.1.4.1 Reduced Integration Method

5.1.4.2 Full Integration Method

5.1.4.3 Improved Integration Method for ALE

5.1.5 Hourglass modes

5.1.5.1 Kosloff & Frasier Formulation [10]

• c is the sound speed,

5.1.5.2 Flanagan-Belytschko Formulation [12]

viIHour = viI − viILin EQ. 5.1.5.3

The linear portion of the velocity field can be expanded to give:

is the hourglass shape vector used in place of ΓIα in EQ. 5.1.5.2.

5.1.5.3 physical hourglass formulation HEPH

In elastic case, you have:

5.1.5.3.1 Advanced elasto-plastic hourglass control

f = σ eq2 (ξ ,η , ζ ) − σ y2 = 0 EQ. 5.1.5.13

1. taking the mean value, i.e.: f ( )

• Check the yield criterion

(σ i )nH+1 = P((σ i )trH

5.1.7 Shock waves

The values of qa and qb are adimensional scalar factors defined as:

5.1.8 Element degeneration

Some examples of element degeneration are shown below.

Not recommended degenerations

5.1.9 Internal stress calculation

The stress rate is comprised of two components:

σɺ ij = σɺ ijv + σɺ ijr EQ. 5.1.9.2

• σɺ ijv is the Jaumann objective stress tensor derivative.

σɺ ijr = σ ik Ω kj + σ jk Ω ki EQ. 5.1.9.3

where Ω is the rigid body rotational velocity tensor (EQ. 2.4.1.11).

σ ij = sij − pδ ij EQ. 5.1.9.4

• sij is the deviatoric stress tensor;

• δ ij is the substitution tensor or unit matrix.

5.1.9.2 Co-rotational Formulation

σɺ ij = σɺ ijv EQ. 5.1.9.5

Isoparametric frames Local (co-rotational)

Fix constant velocity on the top of the

∆σ xxr = ∆Ω y (τ xz + τ zx ) = 2∆Ω yτ xz = α∆Tτ xz EQ. 5.1.9.7

Increment of stress σ zz due to the rigid body motion:

∆σ zzr = −∆Ω y (τ xz + τ zx ) = −2∆Ω yτ xz = −α∆Tτ xz EQ. 5.1.9.8

Increment of shear stress τ xz due to the rigid body motion:

∆τ xzr = ∆Ω y (σ zz − σ xx ) = 2∆Ω yσ zz = α∆Tσ zz EQ. 5.1.9.9

Increment of shear strain:

∆γ xz = ∆T (∂v x / ∂z + ∂v z / ∂x ) = α∆T EQ. 5.1.9.10